#150 by GitHub stars
wonderwhy-er/DesktopCommanderMCP
This is MCP server for Claude that gives it terminal control, file system search and diff file editing capabilities
History
Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.
View weekly changes
plugins/claude/skills/ai-tools-setup/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
plugins/claude/skills/computer-health-check/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
plugins/claude/skills/desktop-commander-overview/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
plugins/claude/skills/knowledge-base/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
plugins/claude/skills/obsidian-vault/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
plugins/claude/skills/terminal/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
plugins/cursor/skills/ai-tools-setup/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
plugins/cursor/skills/computer-health-check/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
plugins/cursor/skills/desktop-commander-overview/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
plugins/cursor/skills/knowledge-base/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
plugins/cursor/skills/obsidian-vault/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
plugins/cursor/skills/terminal/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
skills/ai-tools-setup/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
skills/computer-health-check/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
skills/desktop-commander-overview/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
skills/knowledge-base/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
skills/obsidian-vault/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
skills/terminal/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 9a171d1dae74
No earlier version was observed for comparison.
View on GitHubAdd to my skills
Verified commit 9a171d1dae74de94bf07e6e13878006fb69fa600
git clone --filter=blob:none --no-checkout https://github.com/wonderwhy-er/DesktopCommanderMCP.git .hotskills/wonderwhy-er-desktopcommandermcp && git -C .hotskills/wonderwhy-er-desktopcommandermcp checkout --detach 9a171d1dae74de94bf07e6e13878006fb69fa600
# Copy each selected skill directory from .hotskills/wonderwhy-er-desktopcommandermcp into .claude/skills/<skill-name>/git clone --filter=blob:none --no-checkout https://github.com/wonderwhy-er/DesktopCommanderMCP.git .hotskills/wonderwhy-er-desktopcommandermcp && git -C .hotskills/wonderwhy-er-desktopcommandermcp checkout --detach 9a171d1dae74de94bf07e6e13878006fb69fa600
# Copy each selected skill directory from .hotskills/wonderwhy-er-desktopcommandermcp into .gemini/skills/<skill-name>/10.2kGitHub stars
55stars in 7 days
18skills in this repository
357commits in 30 days
Heat
6.4- Star growth, last 7 daysAhead of 75% of tracked repositories
- Star growth, last 24 hoursAhead of 70% of tracked repositories
- Rank movement, last 7 daysAhead of 55% of tracked repositories
- Commits, last 30 daysAhead of 98% of tracked repositories
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekAhead of 0% of tracked repositories
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
Individual skills in this repository
- ai-tools-setup
plugins/claude/skills/ai-tools-setup - computer-health-check
plugins/claude/skills/computer-health-check - desktop-commander-overview
plugins/claude/skills/desktop-commander-overview - knowledge-base
plugins/claude/skills/knowledge-base - obsidian-vault
plugins/claude/skills/obsidian-vault - terminal
plugins/claude/skills/terminal - ai-tools-setup
plugins/cursor/skills/ai-tools-setup - computer-health-check
plugins/cursor/skills/computer-health-check - desktop-commander-overview
plugins/cursor/skills/desktop-commander-overview - knowledge-base
plugins/cursor/skills/knowledge-base - obsidian-vault
plugins/cursor/skills/obsidian-vault - terminal
plugins/cursor/skills/terminal - ai-tools-setup
skills/ai-tools-setup - computer-health-check
skills/computer-health-check - desktop-commander-overview
skills/desktop-commander-overview - knowledge-base
skills/knowledge-base - obsidian-vault
skills/obsidian-vault - terminal
skills/terminal
How it works
ai-tools-setup
When your agent uses it
Set up, connect, validate, and repair Claude Desktop and MCP servers using Desktop Commander — inspecting and fixing mcp.json / claude_desktop_config.json, getting MCP servers to connect, and diagnosing local AI tooling. Use when a user wants to install, configure, connect, validate, or fix Claude Desktop, an MCP server, an MCP config file, or a local AI agent/gateway. Trigger on the symptoms that actually bring people in: "Claude can't see my tools", "my MCP server isn't showing up", "tools aren't loading in Claude", "add an MCP server", "my MCP isn't connecting", "fix my Claude Desktop config", "the agent won't respond", "wire up Ollama / LM Studio", "restart Claude Desktop". Also covers local agent gateways like OpenClaw and Hermes.
What it loads into context
Frontmatter
206 tokensAlways available to the agent
SKILL.md
1,413 tokensWhen the skill is activated
Reference documents
18,122 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Golden rules (read before doing anything)
- What this skill covers
- General workflow
- A generic "my agent isn't responding" diagnostic
Files
../../../../.claude-plugin/marketplace.jsonasset694 bytes../../.claude-plugin/plugin.jsonasset809 bytesSKILL.mdcore5,845 bytesreferences/claude-desktop-mcp.mddoc3,767 bytesreferences/hermes.mddoc3,902 bytesreferences/openclaw.mddoc3,867 bytes../computer-health-check/SKILL.mddoc8,603 bytes../computer-health-check/references/linux.mddoc3,477 bytes../computer-health-check/references/macos.mddoc3,364 bytes../computer-health-check/references/windows.mddoc4,052 bytes../desktop-commander-overview/SKILL.mddoc12,282 bytes../knowledge-base/SKILL.mddoc8,791 bytes../obsidian-vault/SKILL.mddoc8,950 bytes../terminal/SKILL.mddoc10,761 bytes
Read SKILL.md
# AI Tools Setup Assistant
Help users install, configure, and repair AI tooling on their own machine using
Desktop Commander's file and terminal tools. The work is usually
**configure-and-troubleshoot**, not clean installs — assume something already
exists and may be half-broken.
## Golden rules (read before doing anything)
1. **Detect the OS and shell first.** Read `get_config` and use its `systemInfo`
and `allowedDirectories` — file paths, commands, and config locations differ
across macOS, Windows, and Linux, so confirm the platform before assuming
anything.
2. **Verify the model/API key works before you build.** The most common reasoncomputer-health-check
When your agent uses it
Run a comprehensive, read-only health check on the user's computer and return a scored chat summary with prioritized, plain-English recommendations and safe cleanup suggestions. Use this whenever the user wants to check their computer's health, speed it up, free up / reclaim disk space, find what's eating CPU / memory / storage, check battery health and wear, audit login & startup items, see pending updates, or asks for a "tune-up", "checkup", "system report", "health check", or "is my Mac/PC/laptop healthy". Trigger even on casual phrasing like "my laptop feels slow", "why is my fan so loud", "my computer is laggy", "running out of space", or "clean up my machine". Works on macOS, Windows, and Linux. This is a Desktop Commander skill: it relies on a real local shell (start_process / interact_with_process) and is strictly read-only by default — it never needs sudo and only performs cleanups the user explicitly approves.
What it loads into context
Frontmatter
249 tokensAlways available to the agent
SKILL.md
2,090 tokensWhen the skill is activated
Reference documents
3,267 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What this skill does
- The safety contract (read this first)
- Workflow
- Recommendation rules (findings → advice)
- Notes & pitfalls
Files
SKILL.mdcore8,603 bytesreferences/linux.mddoc3,477 bytesreferences/macos.mddoc3,364 bytesreferences/windows.mddoc4,052 bytes
Read SKILL.md
# Computer Health Check
## What this skill does
Give the user a fast, trustworthy picture of their computer's health and a short list
of actions worth taking — the way a good technician would: look first, explain plainly,
fix only with permission.
It runs entirely through Desktop Commander's local shell, gathers a batch of **read-only**
diagnostics, scores each area, prints a concise summary **in the chat**, and then offers
cleanup suggestions the user can approve one by one.
## The safety contract (read this first)
desktop-commander-overview
When your agent uses it
Use for Desktop Commander MCP capabilities — persistent shells and REPLs, long-running processes, filesystem beyond the workspace, structured files (.xlsx, .docx, .pdf, images) and large local data files such as CSVs, ripgrep search at scale, SSH, or cross-turn state.
What it loads into context
Frontmatter
84 tokensAlways available to the agent
SKILL.md
2,704 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What this MCP gives the agent
- Example workflows
- Core tool inventory
- Conventions
Files
SKILL.mdcore12,282 bytes
Read SKILL.md
# Desktop Commander MCP
Desktop Commander gives the agent reach across the user's actual computer — files, folders, terminals, processes, structured documents, and remote machines reachable over SSH. The tools' detailed schemas (parameters, return shapes, format-specific behavior) live in the MCP itself; this skill explains what they enable and how they compose into common workflows.
## What this MCP gives the agent
**Persistent shell sessions.** Desktop Commander keeps a started process or session alive across tool calls. Inside a single long-lived shell, REPL, or SSH session, state carries forward — environment variables, working directory, activated virtualenvs, open connections, REPL variables — so the agent can `cd`, activate a venv, then send commands or code into that same session many turns later without re-setup. (Note: separate `start_process` calls open separate sessions and do **not** share shell state with each other; persistence is inside one session, not across them.)
**Long-running processes.** Start a dev server, watcher, build, training run, or test suite in the background and keep working. The MCP returns a process handle the agent can tail, interact with, or terminate across many turns. Long-running commands don't need to block the workflow waiting for a foreground command to exit.
**Filesystem reach beyond the IDE workspace.** Read, write, move, list, and inspect files anywhere the user has granted scope — Downloads, Documents, project folders outside the IDE, or any other granted folders. Useful for organize-and-clean tasks, batch document work, and any "look at the file my coworker just sent me" request that doesn't fit inside the IDE sandbox.
**Surgical edits to existing files.** The `edit_block` tool does exact-string find-and-replace with built-in safety: ambiguous matches fail loudly instead of silently overwriting the wrong thing, and an `expected_replacements` count prevents partial-match disasters. Lower data-loss risk than rewriting whole files based on the slice you happened to read — though a wrong `old_string` or wrong `expected_replacements` can still corrupt content, so review the changed content before considering the edit done.
knowledge-base
When your agent uses it
Create and maintain a Markdown knowledge base that any AI agent can read, search, and update. Use when the user wants to start a knowledge base, add or update notes, organize docs/notes for an agent or LLM to consume, build an index of notes, or run a cleanup/maintenance pass on an existing MD knowledge base. Triggers include "knowledge base", "KB", "notes for the agent", "index of notes", "second brain", "docs for AI context", "add a note", "update the KB".
What it loads into context
Frontmatter
144 tokensAlways available to the agent
SKILL.md
2,165 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Core principles (the "why")
- Folder structure (nested by topic)
- Naming rules
- Required frontmatter (every note)
- Note body shape
- INDEX.md (the single source of truth)
- _topic.md (topic map)
- Workflows
- Checklist before finishing any KB edit
Files
SKILL.mdcore8,791 bytesSKILL.mdcore8,791 bytesSKILL.mdcore8,791 bytes
Read SKILL.md
# Knowledge Base (Markdown, agent-readable)
A knowledge base (KB) here is a folder of Markdown files designed so **any** AI
agent can navigate it without a vector database: the agent reads one index file,
picks the relevant notes by their descriptions, and opens only those. Keep the
whole KB readable and the index lean — the index is what gets loaded into
context, so it must be high-signal.
Apply the rules below when creating a new KB, adding/editing notes, or doing a
maintenance pass. When the user's request is ambiguous (new KB vs. add note vs.
cleanup), ask which one before acting.
## Core principles (the "why")
obsidian-vault
When your agent uses it
Organize Obsidian vaults with MOCs, wikilinks, frontmatter/properties, dashboards, orphan-note checks, and cleanup workflows. Use ONLY when the user explicitly refers to Obsidian — by naming "Obsidian", an Obsidian "vault", or an Obsidian-specific feature such as wikilinks ("[[ ]]"), the Dataview or Bases plugins, or Obsidian Properties. Within that Obsidian context it covers creating Maps of Content, adding/fixing wikilinks, finding orphan notes, normalizing YAML frontmatter, generating Dataview/Bases dashboards, organizing folders, deduplicating or renaming notes, and preparing the vault for AI use. Do NOT use this skill for generic note-taking, Markdown, or knowledge-base requests that don't mention Obsidian.
What it loads into context
Frontmatter
207 tokensAlways available to the agent
SKILL.md
2,296 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Wikilinks
- Maps of Content (MOCs)
- Frontmatter / Properties
- Dashboards
- Orphan notes & link hygiene
- Folder organization
- Deduplicate & rename
- Prepare the vault for AI use
- Workflows
- Checklist before finishing
Files
SKILL.mdcore8,950 bytes
Read SKILL.md
# Obsidian Vault Assistant
Help the user organize and maintain an Obsidian vault: build navigation with
MOCs and wikilinks, normalize metadata, generate dashboards, find and fix
orphans, organize folders, deduplicate/rename, and prepare the vault for AI use.
Obsidian-specific facts that shape every rule here:
- Links are **wikilinks** resolved by **filename**, not path: `[[Note title]]`.
Renaming a note auto-updates links inside Obsidian, but moving files outside
Obsidian breaks them — always rename via Obsidian when possible.
- Metadata is **Properties** (YAML frontmatter) at the top of a note.
- Dashboards come from **Dataview** (query language, read-only, flexible) or
**Bases** (native, editable tables, fast on big vaults, properties-only).
terminal
When your agent uses it
Use Desktop Commander for terminal and command-line work, especially anything that needs a shell whose state persists across turns: Python/Node REPLs, database shells, dev servers and other long-running processes, SSH into remote machines, and Windows PowerShell. Also handles everyday terminal tasks — navigating folders, choosing the right command for the user's shell (PowerShell, cmd, bash, zsh), running Docker/curl/cloud-CLI commands, inspecting processes and ports, and saving recurring workflows as scripts — even when the user doesn't say "Desktop Commander" or "terminal." Reach for it on pasted errors like "command not found", "permission denied", "EADDRINUSE", "address already in use", "npm ERR!", "ModuleNotFoundError", or "ENOENT", and on intents like "what's using port 3000", "kill that process", "ssh into my server", or "why won't my dev server start". Works on Windows, macOS, and Linux. Never run destructive commands without explicit confirmation.
What it loads into context
Frontmatter
249 tokensAlways available to the agent
SKILL.md
2,780 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 1. Detect the environment first (do this before suggesting commands)
- 2. Choosing the right command for the shell
- 3. Running commands with Desktop Commander
- 4. Running specific CLIs
- 5. Inspecting processes and ports
- 6. Explaining errors in plain English
- 7. Safety: destructive commands and chaining
- 8. Turning repeated workflows into scripts
- Operating etiquette
Files
SKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytes
Read SKILL.md
# Terminal Command Assistant
Be a calm, safe, cross-platform terminal copilot. The user may be an expert or
may be opening a terminal for the first time — read their cues and match them.
Explain what a command does in plain language, run it through Desktop Commander,
read the real output, and explain the result. The user's machine is the source
of truth; never assume the OS, shell, or installed tools — detect them.
## 1. Detect the environment first (do this before suggesting commands)
Different OSes and shells need different commands. Before recommending or running
anything non-trivial, call **`get_config`** (Desktop Commander) and read
`systemInfo`:
ai-tools-setup
When your agent uses it
Set up, connect, validate, and repair Claude Desktop and MCP servers using Desktop Commander — inspecting and fixing mcp.json / claude_desktop_config.json, getting MCP servers to connect, and diagnosing local AI tooling. Use when a user wants to install, configure, connect, validate, or fix Claude Desktop, an MCP server, an MCP config file, or a local AI agent/gateway. Trigger on the symptoms that actually bring people in: "Claude can't see my tools", "my MCP server isn't showing up", "tools aren't loading in Claude", "add an MCP server", "my MCP isn't connecting", "fix my Claude Desktop config", "the agent won't respond", "wire up Ollama / LM Studio", "restart Claude Desktop". Also covers local agent gateways like OpenClaw and Hermes.
What it loads into context
Frontmatter
206 tokensAlways available to the agent
SKILL.md
1,392 tokensWhen the skill is activated
Reference documents
2,820 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Golden rules (read before doing anything)
- What this skill covers
- General workflow
- A generic "my agent isn't responding" diagnostic
Files
SKILL.mdcore5,759 bytesreferences/claude-desktop-mcp.mddoc3,767 bytesreferences/hermes.mddoc3,902 bytesreferences/openclaw.mddoc3,867 bytes
Read SKILL.md
# AI Tools Setup Assistant
Help users install, configure, and repair AI tooling on their own machine using
Desktop Commander's file and terminal tools. The work is usually
**configure-and-troubleshoot**, not clean installs — assume something already
exists and may be half-broken.
## Golden rules (read before doing anything)
1. **Detect the OS and shell first.** Read `get_config` and use its `systemInfo`
and `allowedDirectories` — file paths, commands, and config locations differ
across macOS, Windows, and Linux, so confirm the platform before assuming
anything.
2. **Verify the model/API key works before you build.** The most common reasoncomputer-health-check
When your agent uses it
Run a comprehensive, read-only health check on the user's computer and return a scored chat summary with prioritized, plain-English recommendations and safe cleanup suggestions. Use this whenever the user wants to check their computer's health, speed it up, free up / reclaim disk space, find what's eating CPU / memory / storage, check battery health and wear, audit login & startup items, see pending updates, or asks for a "tune-up", "checkup", "system report", "health check", or "is my Mac/PC/laptop healthy". Trigger even on casual phrasing like "my laptop feels slow", "why is my fan so loud", "my computer is laggy", "running out of space", or "clean up my machine". Works on macOS, Windows, and Linux. This is a Desktop Commander skill: it relies on a real local shell (start_process / interact_with_process) and is strictly read-only by default — it never needs sudo and only performs cleanups the user explicitly approves.
What it loads into context
Frontmatter
249 tokensAlways available to the agent
SKILL.md
2,077 tokensWhen the skill is activated
Reference documents
3,267 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What this skill does
- The safety contract (read this first)
- Workflow
- Recommendation rules (findings → advice)
- Notes & pitfalls
Files
SKILL.mdcore8,532 bytesreferences/linux.mddoc3,477 bytesreferences/macos.mddoc3,364 bytesreferences/windows.mddoc4,052 bytes
Read SKILL.md
# Computer Health Check
## What this skill does
Give the user a fast, trustworthy picture of their computer's health and a short list
of actions worth taking — the way a good technician would: look first, explain plainly,
fix only with permission.
It runs entirely through Desktop Commander's local shell, gathers a batch of **read-only**
diagnostics, scores each area, prints a concise summary **in the chat**, and then offers
cleanup suggestions the user can approve one by one.
## The safety contract (read this first)
desktop-commander-overview
When your agent uses it
Use for Desktop Commander MCP capabilities — persistent shells and REPLs, long-running processes, filesystem beyond the workspace, structured files (.xlsx, .docx, .pdf, images) and large local data files such as CSVs, ripgrep search at scale, SSH, or cross-turn state.
What it loads into context
Frontmatter
84 tokensAlways available to the agent
SKILL.md
2,704 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What this MCP gives the agent
- Example workflows
- Core tool inventory
- Conventions
Files
SKILL.mdcore12,282 bytes
Read SKILL.md
# Desktop Commander MCP
Desktop Commander gives the agent reach across the user's actual computer — files, folders, terminals, processes, structured documents, and remote machines reachable over SSH. The tools' detailed schemas (parameters, return shapes, format-specific behavior) live in the MCP itself; this skill explains what they enable and how they compose into common workflows.
## What this MCP gives the agent
**Persistent shell sessions.** Desktop Commander keeps a started process or session alive across tool calls. Inside a single long-lived shell, REPL, or SSH session, state carries forward — environment variables, working directory, activated virtualenvs, open connections, REPL variables — so the agent can `cd`, activate a venv, then send commands or code into that same session many turns later without re-setup. (Note: separate `start_process` calls open separate sessions and do **not** share shell state with each other; persistence is inside one session, not across them.)
**Long-running processes.** Start a dev server, watcher, build, training run, or test suite in the background and keep working. The MCP returns a process handle the agent can tail, interact with, or terminate across many turns. Long-running commands don't need to block the workflow waiting for a foreground command to exit.
**Filesystem reach beyond the IDE workspace.** Read, write, move, list, and inspect files anywhere the user has granted scope — Downloads, Documents, project folders outside the IDE, or any other granted folders. Useful for organize-and-clean tasks, batch document work, and any "look at the file my coworker just sent me" request that doesn't fit inside the IDE sandbox.
**Surgical edits to existing files.** The `edit_block` tool does exact-string find-and-replace with built-in safety: ambiguous matches fail loudly instead of silently overwriting the wrong thing, and an `expected_replacements` count prevents partial-match disasters. Lower data-loss risk than rewriting whole files based on the slice you happened to read — though a wrong `old_string` or wrong `expected_replacements` can still corrupt content, so review the changed content before considering the edit done.
knowledge-base
When your agent uses it
Create and maintain a Markdown knowledge base that any AI agent can read, search, and update. Use when the user wants to start a knowledge base, add or update notes, organize docs/notes for an agent or LLM to consume, build an index of notes, or run a cleanup/maintenance pass on an existing MD knowledge base. Triggers include "knowledge base", "KB", "notes for the agent", "index of notes", "second brain", "docs for AI context", "add a note", "update the KB".
What it loads into context
Frontmatter
144 tokensAlways available to the agent
SKILL.md
2,165 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Core principles (the "why")
- Folder structure (nested by topic)
- Naming rules
- Required frontmatter (every note)
- Note body shape
- INDEX.md (the single source of truth)
- _topic.md (topic map)
- Workflows
- Checklist before finishing any KB edit
Files
SKILL.mdcore8,791 bytesSKILL.mdcore8,791 bytesSKILL.mdcore8,791 bytes
Read SKILL.md
# Knowledge Base (Markdown, agent-readable)
A knowledge base (KB) here is a folder of Markdown files designed so **any** AI
agent can navigate it without a vector database: the agent reads one index file,
picks the relevant notes by their descriptions, and opens only those. Keep the
whole KB readable and the index lean — the index is what gets loaded into
context, so it must be high-signal.
Apply the rules below when creating a new KB, adding/editing notes, or doing a
maintenance pass. When the user's request is ambiguous (new KB vs. add note vs.
cleanup), ask which one before acting.
## Core principles (the "why")
obsidian-vault
When your agent uses it
Organize Obsidian vaults with MOCs, wikilinks, frontmatter/properties, dashboards, orphan-note checks, and cleanup workflows. Use ONLY when the user explicitly refers to Obsidian — by naming "Obsidian", an Obsidian "vault", or an Obsidian-specific feature such as wikilinks ("[[ ]]"), the Dataview or Bases plugins, or Obsidian Properties. Within that Obsidian context it covers creating Maps of Content, adding/fixing wikilinks, finding orphan notes, normalizing YAML frontmatter, generating Dataview/Bases dashboards, organizing folders, deduplicating or renaming notes, and preparing the vault for AI use. Do NOT use this skill for generic note-taking, Markdown, or knowledge-base requests that don't mention Obsidian.
What it loads into context
Frontmatter
207 tokensAlways available to the agent
SKILL.md
2,296 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Wikilinks
- Maps of Content (MOCs)
- Frontmatter / Properties
- Dashboards
- Orphan notes & link hygiene
- Folder organization
- Deduplicate & rename
- Prepare the vault for AI use
- Workflows
- Checklist before finishing
Files
SKILL.mdcore8,950 bytes
Read SKILL.md
# Obsidian Vault Assistant
Help the user organize and maintain an Obsidian vault: build navigation with
MOCs and wikilinks, normalize metadata, generate dashboards, find and fix
orphans, organize folders, deduplicate/rename, and prepare the vault for AI use.
Obsidian-specific facts that shape every rule here:
- Links are **wikilinks** resolved by **filename**, not path: `[[Note title]]`.
Renaming a note auto-updates links inside Obsidian, but moving files outside
Obsidian breaks them — always rename via Obsidian when possible.
- Metadata is **Properties** (YAML frontmatter) at the top of a note.
- Dashboards come from **Dataview** (query language, read-only, flexible) or
**Bases** (native, editable tables, fast on big vaults, properties-only).
terminal
When your agent uses it
Use Desktop Commander for terminal and command-line work, especially anything that needs a shell whose state persists across turns: Python/Node REPLs, database shells, dev servers and other long-running processes, SSH into remote machines, and Windows PowerShell. Also handles everyday terminal tasks — navigating folders, choosing the right command for the user's shell (PowerShell, cmd, bash, zsh), running Docker/curl/cloud-CLI commands, inspecting processes and ports, and saving recurring workflows as scripts — even when the user doesn't say "Desktop Commander" or "terminal." Reach for it on pasted errors like "command not found", "permission denied", "EADDRINUSE", "address already in use", "npm ERR!", "ModuleNotFoundError", or "ENOENT", and on intents like "what's using port 3000", "kill that process", "ssh into my server", or "why won't my dev server start". Works on Windows, macOS, and Linux. Never run destructive commands without explicit confirmation.
What it loads into context
Frontmatter
249 tokensAlways available to the agent
SKILL.md
2,780 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 1. Detect the environment first (do this before suggesting commands)
- 2. Choosing the right command for the shell
- 3. Running commands with Desktop Commander
- 4. Running specific CLIs
- 5. Inspecting processes and ports
- 6. Explaining errors in plain English
- 7. Safety: destructive commands and chaining
- 8. Turning repeated workflows into scripts
- Operating etiquette
Files
SKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytes
Read SKILL.md
# Terminal Command Assistant
Be a calm, safe, cross-platform terminal copilot. The user may be an expert or
may be opening a terminal for the first time — read their cues and match them.
Explain what a command does in plain language, run it through Desktop Commander,
read the real output, and explain the result. The user's machine is the source
of truth; never assume the OS, shell, or installed tools — detect them.
## 1. Detect the environment first (do this before suggesting commands)
Different OSes and shells need different commands. Before recommending or running
anything non-trivial, call **`get_config`** (Desktop Commander) and read
`systemInfo`:
ai-tools-setup
When your agent uses it
Set up, connect, validate, and repair Claude Desktop and MCP servers using Desktop Commander — inspecting and fixing mcp.json / claude_desktop_config.json, getting MCP servers to connect, and diagnosing local AI tooling. Use when a user wants to install, configure, connect, validate, or fix Claude Desktop, an MCP server, an MCP config file, or a local AI agent/gateway. Trigger on the symptoms that actually bring people in: "Claude can't see my tools", "my MCP server isn't showing up", "tools aren't loading in Claude", "add an MCP server", "my MCP isn't connecting", "fix my Claude Desktop config", "the agent won't respond", "wire up Ollama / LM Studio", "restart Claude Desktop". Also covers local agent gateways like OpenClaw and Hermes.
What it loads into context
Frontmatter
206 tokensAlways available to the agent
SKILL.md
1,413 tokensWhen the skill is activated
Reference documents
2,820 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Golden rules (read before doing anything)
- What this skill covers
- General workflow
- A generic "my agent isn't responding" diagnostic
Files
SKILL.mdcore5,845 bytesreferences/claude-desktop-mcp.mddoc3,767 bytesreferences/hermes.mddoc3,902 bytesreferences/openclaw.mddoc3,867 bytes
Read SKILL.md
# AI Tools Setup Assistant
Help users install, configure, and repair AI tooling on their own machine using
Desktop Commander's file and terminal tools. The work is usually
**configure-and-troubleshoot**, not clean installs — assume something already
exists and may be half-broken.
## Golden rules (read before doing anything)
1. **Detect the OS and shell first.** Read `get_config` and use its `systemInfo`
and `allowedDirectories` — file paths, commands, and config locations differ
across macOS, Windows, and Linux, so confirm the platform before assuming
anything.
2. **Verify the model/API key works before you build.** The most common reasoncomputer-health-check
When your agent uses it
Run a comprehensive, read-only health check on the user's computer and return a scored chat summary with prioritized, plain-English recommendations and safe cleanup suggestions. Use this whenever the user wants to check their computer's health, speed it up, free up / reclaim disk space, find what's eating CPU / memory / storage, check battery health and wear, audit login & startup items, see pending updates, or asks for a "tune-up", "checkup", "system report", "health check", or "is my Mac/PC/laptop healthy". Trigger even on casual phrasing like "my laptop feels slow", "why is my fan so loud", "my computer is laggy", "running out of space", or "clean up my machine". Works on macOS, Windows, and Linux. This is a Desktop Commander skill: it relies on a real local shell (start_process / interact_with_process) and is strictly read-only by default — it never needs sudo and only performs cleanups the user explicitly approves.
What it loads into context
Frontmatter
249 tokensAlways available to the agent
SKILL.md
2,090 tokensWhen the skill is activated
Reference documents
3,267 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What this skill does
- The safety contract (read this first)
- Workflow
- Recommendation rules (findings → advice)
- Notes & pitfalls
Files
SKILL.mdcore8,603 bytesreferences/linux.mddoc3,477 bytesreferences/macos.mddoc3,364 bytesreferences/windows.mddoc4,052 bytes
Read SKILL.md
# Computer Health Check
## What this skill does
Give the user a fast, trustworthy picture of their computer's health and a short list
of actions worth taking — the way a good technician would: look first, explain plainly,
fix only with permission.
It runs entirely through Desktop Commander's local shell, gathers a batch of **read-only**
diagnostics, scores each area, prints a concise summary **in the chat**, and then offers
cleanup suggestions the user can approve one by one.
## The safety contract (read this first)
desktop-commander-overview
When your agent uses it
Use for Desktop Commander MCP capabilities — persistent shells and REPLs, long-running processes, filesystem beyond the workspace, structured files (.xlsx, .docx, .pdf, images) and large local data files such as CSVs, ripgrep search at scale, SSH, or cross-turn state.
What it loads into context
Frontmatter
84 tokensAlways available to the agent
SKILL.md
2,704 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- What this MCP gives the agent
- Example workflows
- Core tool inventory
- Conventions
Files
SKILL.mdcore12,282 bytes
Read SKILL.md
# Desktop Commander MCP
Desktop Commander gives the agent reach across the user's actual computer — files, folders, terminals, processes, structured documents, and remote machines reachable over SSH. The tools' detailed schemas (parameters, return shapes, format-specific behavior) live in the MCP itself; this skill explains what they enable and how they compose into common workflows.
## What this MCP gives the agent
**Persistent shell sessions.** Desktop Commander keeps a started process or session alive across tool calls. Inside a single long-lived shell, REPL, or SSH session, state carries forward — environment variables, working directory, activated virtualenvs, open connections, REPL variables — so the agent can `cd`, activate a venv, then send commands or code into that same session many turns later without re-setup. (Note: separate `start_process` calls open separate sessions and do **not** share shell state with each other; persistence is inside one session, not across them.)
**Long-running processes.** Start a dev server, watcher, build, training run, or test suite in the background and keep working. The MCP returns a process handle the agent can tail, interact with, or terminate across many turns. Long-running commands don't need to block the workflow waiting for a foreground command to exit.
**Filesystem reach beyond the IDE workspace.** Read, write, move, list, and inspect files anywhere the user has granted scope — Downloads, Documents, project folders outside the IDE, or any other granted folders. Useful for organize-and-clean tasks, batch document work, and any "look at the file my coworker just sent me" request that doesn't fit inside the IDE sandbox.
**Surgical edits to existing files.** The `edit_block` tool does exact-string find-and-replace with built-in safety: ambiguous matches fail loudly instead of silently overwriting the wrong thing, and an `expected_replacements` count prevents partial-match disasters. Lower data-loss risk than rewriting whole files based on the slice you happened to read — though a wrong `old_string` or wrong `expected_replacements` can still corrupt content, so review the changed content before considering the edit done.
knowledge-base
When your agent uses it
Create and maintain a Markdown knowledge base that any AI agent can read, search, and update. Use when the user wants to start a knowledge base, add or update notes, organize docs/notes for an agent or LLM to consume, build an index of notes, or run a cleanup/maintenance pass on an existing MD knowledge base. Triggers include "knowledge base", "KB", "notes for the agent", "index of notes", "second brain", "docs for AI context", "add a note", "update the KB".
What it loads into context
Frontmatter
144 tokensAlways available to the agent
SKILL.md
2,165 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Core principles (the "why")
- Folder structure (nested by topic)
- Naming rules
- Required frontmatter (every note)
- Note body shape
- INDEX.md (the single source of truth)
- _topic.md (topic map)
- Workflows
- Checklist before finishing any KB edit
Files
SKILL.mdcore8,791 bytesSKILL.mdcore8,791 bytesSKILL.mdcore8,791 bytes
Read SKILL.md
# Knowledge Base (Markdown, agent-readable)
A knowledge base (KB) here is a folder of Markdown files designed so **any** AI
agent can navigate it without a vector database: the agent reads one index file,
picks the relevant notes by their descriptions, and opens only those. Keep the
whole KB readable and the index lean — the index is what gets loaded into
context, so it must be high-signal.
Apply the rules below when creating a new KB, adding/editing notes, or doing a
maintenance pass. When the user's request is ambiguous (new KB vs. add note vs.
cleanup), ask which one before acting.
## Core principles (the "why")
obsidian-vault
When your agent uses it
Organize Obsidian vaults with MOCs, wikilinks, frontmatter/properties, dashboards, orphan-note checks, and cleanup workflows. Use ONLY when the user explicitly refers to Obsidian — by naming "Obsidian", an Obsidian "vault", or an Obsidian-specific feature such as wikilinks ("[[ ]]"), the Dataview or Bases plugins, or Obsidian Properties. Within that Obsidian context it covers creating Maps of Content, adding/fixing wikilinks, finding orphan notes, normalizing YAML frontmatter, generating Dataview/Bases dashboards, organizing folders, deduplicating or renaming notes, and preparing the vault for AI use. Do NOT use this skill for generic note-taking, Markdown, or knowledge-base requests that don't mention Obsidian.
What it loads into context
Frontmatter
207 tokensAlways available to the agent
SKILL.md
2,296 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Wikilinks
- Maps of Content (MOCs)
- Frontmatter / Properties
- Dashboards
- Orphan notes & link hygiene
- Folder organization
- Deduplicate & rename
- Prepare the vault for AI use
- Workflows
- Checklist before finishing
Files
SKILL.mdcore8,950 bytes
Read SKILL.md
# Obsidian Vault Assistant
Help the user organize and maintain an Obsidian vault: build navigation with
MOCs and wikilinks, normalize metadata, generate dashboards, find and fix
orphans, organize folders, deduplicate/rename, and prepare the vault for AI use.
Obsidian-specific facts that shape every rule here:
- Links are **wikilinks** resolved by **filename**, not path: `[[Note title]]`.
Renaming a note auto-updates links inside Obsidian, but moving files outside
Obsidian breaks them — always rename via Obsidian when possible.
- Metadata is **Properties** (YAML frontmatter) at the top of a note.
- Dashboards come from **Dataview** (query language, read-only, flexible) or
**Bases** (native, editable tables, fast on big vaults, properties-only).
terminal
When your agent uses it
Use Desktop Commander for terminal and command-line work, especially anything that needs a shell whose state persists across turns: Python/Node REPLs, database shells, dev servers and other long-running processes, SSH into remote machines, and Windows PowerShell. Also handles everyday terminal tasks — navigating folders, choosing the right command for the user's shell (PowerShell, cmd, bash, zsh), running Docker/curl/cloud-CLI commands, inspecting processes and ports, and saving recurring workflows as scripts — even when the user doesn't say "Desktop Commander" or "terminal." Reach for it on pasted errors like "command not found", "permission denied", "EADDRINUSE", "address already in use", "npm ERR!", "ModuleNotFoundError", or "ENOENT", and on intents like "what's using port 3000", "kill that process", "ssh into my server", or "why won't my dev server start". Works on Windows, macOS, and Linux. Never run destructive commands without explicit confirmation.
What it loads into context
Frontmatter
249 tokensAlways available to the agent
SKILL.md
2,780 tokensWhen the skill is activated
Reference documents
0 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- 1. Detect the environment first (do this before suggesting commands)
- 2. Choosing the right command for the shell
- 3. Running commands with Desktop Commander
- 4. Running specific CLIs
- 5. Inspecting processes and ports
- 6. Explaining errors in plain English
- 7. Safety: destructive commands and chaining
- 8. Turning repeated workflows into scripts
- Operating etiquette
Files
SKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytesSKILL.mdcore10,761 bytes
Read SKILL.md
# Terminal Command Assistant
Be a calm, safe, cross-platform terminal copilot. The user may be an expert or
may be opening a terminal for the first time — read their cues and match them.
Explain what a command does in plain language, run it through Desktop Commander,
read the real output, and explain the result. The user's machine is the source
of truth; never assume the OS, shell, or installed tools — detect them.
## 1. Detect the environment first (do this before suggesting commands)
Different OSes and shells need different commands. Before recommending or running
anything non-trivial, call **`get_config`** (Desktop Commander) and read
`systemInfo`:
Skills included
- ai-tools-setup
- computer-health-check
- desktop-commander-overview
- knowledge-base
- obsidian-vault
- terminal
- ai-tools-setup
- computer-health-check
- desktop-commander-overview
- knowledge-base
- obsidian-vault
- terminal
- ai-tools-setup
- computer-health-check
- desktop-commander-overview
- knowledge-base
- obsidian-vault
- terminal
Repository facts
- License
- MIT
- Created
- Dec 4, 2024
- Last pushed
- Nov 5, 2026
- Forks
- 1,276
- Open issues
- 209
- Project activity
- Active
Rank over the last 12 weeks
Now #150 · best #126 · down 12 in 30 days
- #145 on Oct 29
- #146 from Oct 30 to Oct 31
- #147 on Nov 1
- #150 on Nov 2
- #151 on Nov 3
- #150 since Nov 4
49 earlier periods
- #124 on Aug 2
- #125 on Aug 7
- #123 on Aug 8
- #124 on Aug 9
- #126 from Aug 10 to Aug 11
- #127 from Aug 12 to Aug 15
- #126 on Aug 16
- #127 from Aug 17 to Aug 21
- #126 on Aug 22
- #127 from Aug 23 to Aug 25
- #128 from Aug 26 to Aug 29
- #127 on Aug 30
- #128 on Aug 31
- #129 on Sep 1
- #128 on Sep 2
- #130 on Sep 3
- #129 on Sep 4
- #130 on Sep 5
- #129 on Sep 6
- #130 from Sep 7 to Sep 10
- #129 on Sep 11
- #130 from Sep 12 to Sep 13
- #131 on Sep 14
- #132 on Sep 15
- #131 on Sep 16
- #132 on Sep 17
- #134 on Sep 18
- #133 on Sep 19
- #134 from Sep 20 to Sep 23
- #133 on Sep 24
- #134 from Sep 25 to Sep 27
- #136 from Sep 28 to Sep 29
- #135 on Sep 30
- #136 from Oct 1 to Oct 2
- #135 from Oct 3 to Oct 4
- #138 on Oct 5
- #136 on Oct 6
- #138 from Oct 7 to Oct 9
- #139 from Oct 10 to Oct 13
- #140 on Oct 14
- #139 on Oct 15
- #138 on Oct 16
- #140 on Oct 17
- #141 on Oct 18
- #142 on Oct 19
- #143 from Oct 20 to Oct 22
- #144 from Oct 23 to Oct 24
- #145 on Oct 25
- #144 on Oct 28
Before you install
Network or broad shellAutomated signals, not a guarantee
Checked Nov 2, 2026
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 21 notes
plugins/claude/skills/knowledge-base/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/claude/skills/knowledge-base/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/claude/skills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/claude/skills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/claude/skills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/claude/skills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/claude/skills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/cursor/skills/knowledge-base/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/cursor/skills/knowledge-base/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/cursor/skills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/cursor/skills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/cursor/skills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/cursor/skills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/cursor/skills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/knowledge-base/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/knowledge-base/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedskills/terminal/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- 35 dangerous shell commandsin 15 files: plugins/claude/skills/terminal/SKILL.md:58 plugins/claude/skills/computer-health-check/references/linux.md:35 terminal/SKILL.md:58 and 12 more
- 108 network referencesin 16 files: plugins/claude/skills/terminal/SKILL.md:108 ai-tools-setup/references/hermes.md:6 ai-tools-setup/references/openclaw.md:3 and 13 more
- No hidden or obfuscated text
- No instruction overrides
- No executable scripts
- No requests to vote in Hot
Show all 146 signals by file
plugins/claude/.claude-plugin/plugin.json
- Network accessline 8
"email": "es@desktopcommander.app" }, "homepage": "https://desktopcommander.app", - Network accessline 9
}, "homepage": "https://desktopcommander.app", "repository": "https://github.com/wonderwhy-er/DesktopCommanderMCP", - Network accessline 10
"homepage": "https://desktopcommander.app", - Network accessline 11
"repository": "https://github.com/wonderwhy-er/DesktopCommanderMCP",
plugins/claude/skills/ai-tools-setup/SKILL.md
- Network accessline 32
these sessions dead-end is a provider auth or billing error discovered *after* a lot of setup work. Make one tiny test call with `start_process` (a quick curl to the provider) early, and tell the user plainly if the key is invalid - Network accessline 33
*after* a lot of setup work. Make one tiny test call with `start_process` (a quick curl to the provider) early, and tell the user plainly if the key is invalid or out of credit. - Network accessline 34
curl to the provider) early, and tell the user plainly if the key is invalid - Network accessline 79
When in doubt about a tool's current behavior, fetch the official docs: OpenClaw — https://docs.openclaw.ai/ · - Network accessline 80
OpenClaw — https://docs.openclaw.ai/ · - Network accessline 81
Hermes — https://hermes-agent.nousresearch.com/docs/
plugins/claude/skills/ai-tools-setup/references/claude-desktop-mcp.md
- Access to secretsline 34
[Credential-related source omitted]
plugins/claude/skills/ai-tools-setup/references/hermes.md
- Network accessline 6
Research (a learning loop that creates/improves skills, persistent memory, runs across many platforms). This is almost always what users mean when they say "set up Hermes." Docs: https://hermes-agent.nousresearch.com/docs/ - Network accessline 7
runs across many platforms). This is almost always what users mean when they say "set up Hermes." Docs: https://hermes-agent.nousresearch.com/docs/ 2. **Hermes 3 (and other Hermes models)** — Nous Research's open-weight LLMs. - Network accessline 8
say "set up Hermes." Docs: https://hermes-agent.nousresearch.com/docs/ - Network accessline 19
**Desktop app (Windows/macOS):** download the Hermes Desktop installer from https://hermes-agent.nousresearch.com/ and run it (installs CLI + desktop app). - Network accessline 20
https://hermes-agent.nousresearch.com/ and run it (installs CLI + desktop app). - Network accessline 25
# Linux / macOS / WSL2 / Android (Termux) curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash - Dangerous shell operationline 26
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash - Network accessline 26
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash - Network accessline 30
# Windows (native), in PowerShell iex (irm https://hermes-agent.nousresearch.com/install.ps1) - Network accessline 31
iex (irm https://hermes-agent.nousresearch.com/install.ps1)
plugins/claude/skills/ai-tools-setup/references/openclaw.md
- Network accessline 3
OpenClaw is a self-hosted, multi-channel **gateway for AI agents** — one Gateway process bridges chat apps (Discord, iMessage, Signal, Slack, Telegram, WhatsApp, and more) to an AI agent. MIT licensed. Official docs: https://docs.openclaw.a - Network accessline 4
process bridges chat apps (Discord, iMessage, Signal, Slack, Telegram, WhatsApp, and more) to an AI agent. MIT licensed. Official docs: https://docs.openclaw.ai/ (append `/index.md` or use the "View as Markdown" links for clean text). - Network accessline 5
and more) to an AI agent. MIT licensed. Official docs: https://docs.openclaw.ai/ - Network accessline 26
The Control UI default is `http://127.0.0.1:18789/`. (Some versions/setups use a - Network accessline 72
Each chat platform is a "channel" with its own setup (token/pairing). Telegram is the quickest to validate. Configure one channel, confirm a round-trip message, then add others. See https://docs.openclaw.ai/channels. - Network accessline 74
message, then add others. See https://docs.openclaw.ai/channels. - Network accessline 78
To reach the gateway from outside the machine, use the documented patterns (SSH / tailnet). Tailscale is covered at https://docs.openclaw.ai/gateway/tailscale. Avoid exposing the dashboard - Network accessline 79
(SSH / tailnet). Tailscale is covered at https://docs.openclaw.ai/gateway/tailscale. Avoid exposing the dashboard directly to the public internet. - Network accessline 80
https://docs.openclaw.ai/gateway/tailscale. Avoid exposing the dashboard
plugins/claude/skills/computer-health-check/SKILL.md
plugins/claude/skills/computer-health-check/references/linux.md
- Dangerous shell operationline 35
- **Disk health:** `lsblk` confirms devices; full SMART needs `sudo smartctl -H /dev/sdX` (skip — - Dangerous shell operationline 35
- **Disk health:** `lsblk` confirms devices; full SMART needs `sudo smartctl -H /dev/sdX` (skip — - Dangerous shell operationline 50
- **Empty Trash (safe):** `rm -rf "$HOME/.local/share/Trash/"*`. - Dangerous shell operationline 50
- **Empty Trash (safe):** `rm -rf "$HOME/.local/share/Trash/"*`.
plugins/claude/skills/computer-health-check/references/macos.md
plugins/claude/skills/desktop-commander-overview/SKILL.md
- Network accessline 48
`start_process("npm run dev", timeout_ms=...)` keeps the server up. The agent then loops: `edit_block` on the route file, `read_process_output(pid, offset=-30)` to see the server's reload, `start_process("curl -s http://localhost:3000/api/. - Network accessline 48
`start_process("npm run dev", timeout_ms=...)` keeps the server up. The agent then loops: `edit_block` on the route file, `read_process_output(pid, offset=-30)` to see the server's reload, `start_process("curl -s http://localhost:3000/api/.
plugins/claude/skills/terminal/SKILL.md
- Dangerous shell operationline 58
| Delete file | `rm file` | `Remove-Item file` | `del file` | - Dangerous shell operationline 58
| Delete file | `rm file` | `Remove-Item file` | `del file` | - Network accessline 108
`interact_with_process`. Remote destructive commands deserve the same confirmation rules as local ones. - **curl / HTTP**: fine for inspection (`curl -i https://...`). Be careful with - Network accessline 108
`interact_with_process`. Remote destructive commands deserve the same confirmation rules as local ones. - **curl / HTTP**: fine for inspection (`curl -i https://...`). Be careful with - Network accessline 109
confirmation rules as local ones. - **curl / HTTP**: fine for inspection (`curl -i https://...`). Be careful with anything that pipes a downloaded script straight into a shell - Network accessline 109
confirmation rules as local ones. - **curl / HTTP**: fine for inspection (`curl -i https://...`). Be careful with anything that pipes a downloaded script straight into a shell - Network accessline 110
- **curl / HTTP**: fine for inspection (`curl -i https://...`). Be careful with - Network accessline 110
- **curl / HTTP**: fine for inspection (`curl -i https://...`). Be careful with - Network accessline 111
anything that pipes a downloaded script straight into a shell (`curl ... | sh`) — that's untrusted code execution; show it and confirm first. - **Cloud CLIs** (`aws`, `gcloud`, `az`): read-only commands (`describe`, `list`, - Network accessline 111
anything that pipes a downloaded script straight into a shell (`curl ... | sh`) — that's untrusted code execution; show it and confirm first. - **Cloud CLIs** (`aws`, `gcloud`, `az`): read-only commands (`describe`, `list`, - Dangerous shell operationline 112
(`curl ... | sh`) — that's untrusted code execution; show it and confirm first. - Dangerous shell operationline 112
(`curl ... | sh`) — that's untrusted code execution; show it and confirm first. - Network accessline 112
(`curl ... | sh`) — that's untrusted code execution; show it and confirm first. - Network accessline 112
(`curl ... | sh`) — that's untrusted code execution; show it and confirm first. - Dangerous shell operationline 158
- Recursive/forced deletes: `rm -rf`, `Remove-Item -Recurse -Force`, `del /s`, - Dangerous shell operationline 158
- Recursive/forced deletes: `rm -rf`, `Remove-Item -Recurse -Force`, `del /s`, - Network accessline 164
- `chmod -R` / `chown -R` on broad paths, `killall`, mass process kills. - Cloud deletes/scaling/IAM changes; `docker ... prune`/`rm`/volume removal. - Anything piping remote content into a shell (`curl ... | sh`). - Network accessline 164
- `chmod -R` / `chown -R` on broad paths, `killall`, mass process kills. - Cloud deletes/scaling/IAM changes; `docker ... prune`/`rm`/volume removal. - Anything piping remote content into a shell (`curl ... | sh`). - Dangerous shell operationline 166
- Anything piping remote content into a shell (`curl ... | sh`). - Dangerous shell operationline 166
- Anything piping remote content into a shell (`curl ... | sh`). - Network accessline 166
- Anything piping remote content into a shell (`curl ... | sh`). - Network accessline 166
- Anything piping remote content into a shell (`curl ... | sh`).
plugins/cursor/skills/ai-tools-setup/SKILL.md
- Network accessline 32
these sessions dead-end is a provider auth or billing error discovered *after* a lot of setup work. Make one tiny test call with `start_process` (a quick curl to the provider) early, and tell the user plainly if the key is invalid - Network accessline 33
*after* a lot of setup work. Make one tiny test call with `start_process` (a quick curl to the provider) early, and tell the user plainly if the key is invalid or out of credit. - Network accessline 34
curl to the provider) early, and tell the user plainly if the key is invalid - Network accessline 79
When in doubt about a tool's current behavior, fetch the official docs: OpenClaw — https://docs.openclaw.ai/ · - Network accessline 80
OpenClaw — https://docs.openclaw.ai/ · - Network accessline 81
Hermes — https://hermes-agent.nousresearch.com/docs/
plugins/cursor/skills/ai-tools-setup/references/claude-desktop-mcp.md
- Access to secretsline 34
[Credential-related source omitted]
plugins/cursor/skills/ai-tools-setup/references/hermes.md
- Network accessline 6
Research (a learning loop that creates/improves skills, persistent memory, runs across many platforms). This is almost always what users mean when they say "set up Hermes." Docs: https://hermes-agent.nousresearch.com/docs/ - Network accessline 7
runs across many platforms). This is almost always what users mean when they say "set up Hermes." Docs: https://hermes-agent.nousresearch.com/docs/ 2. **Hermes 3 (and other Hermes models)** — Nous Research's open-weight LLMs. - Network accessline 8
say "set up Hermes." Docs: https://hermes-agent.nousresearch.com/docs/ - Network accessline 19
**Desktop app (Windows/macOS):** download the Hermes Desktop installer from https://hermes-agent.nousresearch.com/ and run it (installs CLI + desktop app). - Network accessline 20
https://hermes-agent.nousresearch.com/ and run it (installs CLI + desktop app). - Network accessline 25
# Linux / macOS / WSL2 / Android (Termux) curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash - Dangerous shell operationline 26
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash - Network accessline 26
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash - Network accessline 30
# Windows (native), in PowerShell iex (irm https://hermes-agent.nousresearch.com/install.ps1) - Network accessline 31
iex (irm https://hermes-agent.nousresearch.com/install.ps1)
plugins/cursor/skills/ai-tools-setup/references/openclaw.md
- Network accessline 3
OpenClaw is a self-hosted, multi-channel **gateway for AI agents** — one Gateway process bridges chat apps (Discord, iMessage, Signal, Slack, Telegram, WhatsApp, and more) to an AI agent. MIT licensed. Official docs: https://docs.openclaw.a - Network accessline 4
process bridges chat apps (Discord, iMessage, Signal, Slack, Telegram, WhatsApp, and more) to an AI agent. MIT licensed. Official docs: https://docs.openclaw.ai/ (append `/index.md` or use the "View as Markdown" links for clean text). - Network accessline 5
and more) to an AI agent. MIT licensed. Official docs: https://docs.openclaw.ai/ - Network accessline 26
The Control UI default is `http://127.0.0.1:18789/`. (Some versions/setups use a - Network accessline 72
Each chat platform is a "channel" with its own setup (token/pairing). Telegram is the quickest to validate. Configure one channel, confirm a round-trip message, then add others. See https://docs.openclaw.ai/channels. - Network accessline 74
message, then add others. See https://docs.openclaw.ai/channels. - Network accessline 78
To reach the gateway from outside the machine, use the documented patterns (SSH / tailnet). Tailscale is covered at https://docs.openclaw.ai/gateway/tailscale. Avoid exposing the dashboard - Network accessline 79
(SSH / tailnet). Tailscale is covered at https://docs.openclaw.ai/gateway/tailscale. Avoid exposing the dashboard directly to the public internet. - Network accessline 80
https://docs.openclaw.ai/gateway/tailscale. Avoid exposing the dashboard
plugins/cursor/skills/computer-health-check/SKILL.md
- Dangerous shell operationline 16
default — it never needs sudo and only performs cleanups the user explicitly approves.
plugins/cursor/skills/computer-health-check/references/linux.md
plugins/cursor/skills/computer-health-check/references/macos.md
- Dangerous shell operationline 54
- **Empty Trash (safe):** `rm -rf "$HOME/.Trash/"*` — or tell the user to empty it from Finder.
plugins/cursor/skills/desktop-commander-overview/SKILL.md
- Network accessline 48
`start_process("npm run dev", timeout_ms=...)` keeps the server up. The agent then loops: `edit_block` on the route file, `read_process_output(pid, offset=-30)` to see the server's reload, `start_process("curl -s http://localhost:3000/api/.
plugins/cursor/skills/terminal/SKILL.md
- Dangerous shell operationline 58
| Delete file | `rm file` | `Remove-Item file` | `del file` | - Network accessline 108
`interact_with_process`. Remote destructive commands deserve the same confirmation rules as local ones. - **curl / HTTP**: fine for inspection (`curl -i https://...`). Be careful with - Network accessline 109
confirmation rules as local ones. - **curl / HTTP**: fine for inspection (`curl -i https://...`). Be careful with anything that pipes a downloaded script straight into a shell - Network accessline 110
- **curl / HTTP**: fine for inspection (`curl -i https://...`). Be careful with - Network accessline 111
anything that pipes a downloaded script straight into a shell (`curl ... | sh`) — that's untrusted code execution; show it and confirm first. - **Cloud CLIs** (`aws`, `gcloud`, `az`): read-only commands (`describe`, `list`, - Dangerous shell operationline 112
(`curl ... | sh`) — that's untrusted code execution; show it and confirm first. - Network accessline 112
(`curl ... | sh`) — that's untrusted code execution; show it and confirm first. - Dangerous shell operationline 158
- Recursive/forced deletes: `rm -rf`, `Remove-Item -Recurse -Force`, `del /s`, - Network accessline 164
- `chmod -R` / `chown -R` on broad paths, `killall`, mass process kills. - Cloud deletes/scaling/IAM changes; `docker ... prune`/`rm`/volume removal. - Anything piping remote content into a shell (`curl ... | sh`). - Dangerous shell operationline 166
- Anything piping remote content into a shell (`curl ... | sh`). - Network accessline 166
- Anything piping remote content into a shell (`curl ... | sh`).
skills/ai-tools-setup/SKILL.md
- Network accessline 32
these sessions dead-end is a provider auth or billing error discovered *after* a lot of setup work. Make one tiny test call with `start_process` (a quick curl to the provider) early, and tell the user plainly if the key is invalid - Network accessline 33
*after* a lot of setup work. Make one tiny test call with `start_process` (a quick curl to the provider) early, and tell the user plainly if the key is invalid or out of credit. - Network accessline 34
curl to the provider) early, and tell the user plainly if the key is invalid - Network accessline 79
When in doubt about a tool's current behavior, fetch the official docs: OpenClaw — https://docs.openclaw.ai/ · - Network accessline 80
OpenClaw — https://docs.openclaw.ai/ · - Network accessline 81
Hermes — https://hermes-agent.nousresearch.com/docs/
skills/ai-tools-setup/references/claude-desktop-mcp.md
- Access to secretsline 34
[Credential-related source omitted]
skills/ai-tools-setup/references/hermes.md
- Network accessline 6
Research (a learning loop that creates/improves skills, persistent memory, runs across many platforms). This is almost always what users mean when they say "set up Hermes." Docs: https://hermes-agent.nousresearch.com/docs/ - Network accessline 7
runs across many platforms). This is almost always what users mean when they say "set up Hermes." Docs: https://hermes-agent.nousresearch.com/docs/ 2. **Hermes 3 (and other Hermes models)** — Nous Research's open-weight LLMs. - Network accessline 8
say "set up Hermes." Docs: https://hermes-agent.nousresearch.com/docs/ - Network accessline 19
**Desktop app (Windows/macOS):** download the Hermes Desktop installer from https://hermes-agent.nousresearch.com/ and run it (installs CLI + desktop app). - Network accessline 20
https://hermes-agent.nousresearch.com/ and run it (installs CLI + desktop app). - Network accessline 25
# Linux / macOS / WSL2 / Android (Termux) curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash - Dangerous shell operationline 26
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash - Network accessline 26
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash - Network accessline 30
# Windows (native), in PowerShell iex (irm https://hermes-agent.nousresearch.com/install.ps1) - Network accessline 31
iex (irm https://hermes-agent.nousresearch.com/install.ps1)
skills/ai-tools-setup/references/openclaw.md
- Network accessline 3
OpenClaw is a self-hosted, multi-channel **gateway for AI agents** — one Gateway process bridges chat apps (Discord, iMessage, Signal, Slack, Telegram, WhatsApp, and more) to an AI agent. MIT licensed. Official docs: https://docs.openclaw.a - Network accessline 4
process bridges chat apps (Discord, iMessage, Signal, Slack, Telegram, WhatsApp, and more) to an AI agent. MIT licensed. Official docs: https://docs.openclaw.ai/ (append `/index.md` or use the "View as Markdown" links for clean text). - Network accessline 5
and more) to an AI agent. MIT licensed. Official docs: https://docs.openclaw.ai/ - Network accessline 26
The Control UI default is `http://127.0.0.1:18789/`. (Some versions/setups use a - Network accessline 72
Each chat platform is a "channel" with its own setup (token/pairing). Telegram is the quickest to validate. Configure one channel, confirm a round-trip message, then add others. See https://docs.openclaw.ai/channels. - Network accessline 74
message, then add others. See https://docs.openclaw.ai/channels. - Network accessline 78
To reach the gateway from outside the machine, use the documented patterns (SSH / tailnet). Tailscale is covered at https://docs.openclaw.ai/gateway/tailscale. Avoid exposing the dashboard - Network accessline 79
(SSH / tailnet). Tailscale is covered at https://docs.openclaw.ai/gateway/tailscale. Avoid exposing the dashboard directly to the public internet. - Network accessline 80
https://docs.openclaw.ai/gateway/tailscale. Avoid exposing the dashboard
skills/computer-health-check/SKILL.md
- Dangerous shell operationline 16
default — it never needs sudo and only performs cleanups the user explicitly approves.
skills/computer-health-check/references/linux.md
skills/computer-health-check/references/macos.md
- Dangerous shell operationline 54
- **Empty Trash (safe):** `rm -rf "$HOME/.Trash/"*` — or tell the user to empty it from Finder.
skills/desktop-commander-overview/SKILL.md
- Network accessline 48
`start_process("npm run dev", timeout_ms=...)` keeps the server up. The agent then loops: `edit_block` on the route file, `read_process_output(pid, offset=-30)` to see the server's reload, `start_process("curl -s http://localhost:3000/api/.
skills/terminal/SKILL.md
- Dangerous shell operationline 58
| Delete file | `rm file` | `Remove-Item file` | `del file` | - Network accessline 108
`interact_with_process`. Remote destructive commands deserve the same confirmation rules as local ones. - **curl / HTTP**: fine for inspection (`curl -i https://...`). Be careful with - Network accessline 109
confirmation rules as local ones. - **curl / HTTP**: fine for inspection (`curl -i https://...`). Be careful with anything that pipes a downloaded script straight into a shell - Network accessline 110
- **curl / HTTP**: fine for inspection (`curl -i https://...`). Be careful with - Network accessline 111
anything that pipes a downloaded script straight into a shell (`curl ... | sh`) — that's untrusted code execution; show it and confirm first. - **Cloud CLIs** (`aws`, `gcloud`, `az`): read-only commands (`describe`, `list`, - Dangerous shell operationline 112
(`curl ... | sh`) — that's untrusted code execution; show it and confirm first. - Network accessline 112
(`curl ... | sh`) — that's untrusted code execution; show it and confirm first. - Dangerous shell operationline 158
- Recursive/forced deletes: `rm -rf`, `Remove-Item -Recurse -Force`, `del /s`, - Network accessline 164
- `chmod -R` / `chown -R` on broad paths, `killall`, mass process kills. - Cloud deletes/scaling/IAM changes; `docker ... prune`/`rm`/volume removal. - Anything piping remote content into a shell (`curl ... | sh`). - Dangerous shell operationline 166
- Anything piping remote content into a shell (`curl ... | sh`). - Network accessline 166
- Anything piping remote content into a shell (`curl ... | sh`).
How the signals fit together
Facts from a static read of the files at the pinned commit. They describe what the code does, not why.
Some files could not be read; the facts below still hold.
Signal confidence: 142 facts
- Network access in prose of the SKILL.mdai-tools-setup/SKILL.md:32Low confidence
- Network access in prose of the SKILL.mdai-tools-setup/SKILL.md:33Low confidence
- Network access in prose of the SKILL.mdai-tools-setup/SKILL.md:34Low confidence
- Network access in prose of the SKILL.mdai-tools-setup/SKILL.md:79Low confidence
- Network access in prose of the SKILL.mdai-tools-setup/SKILL.md:80Low confidence
- Network access in prose of the SKILL.mdai-tools-setup/SKILL.md:81Low confidence
- Access to secrets in code of documentationai-tools-setup/references/claude-desktop-mcp.md:34Low confidence
- Network access in prose of documentationai-tools-setup/references/hermes.md:6Low confidence
- Network access in prose of documentationai-tools-setup/references/hermes.md:7Low confidence
- Network access in prose of documentationai-tools-setup/references/hermes.md:8Low confidence
- Network access in prose of documentationai-tools-setup/references/hermes.md:19Low confidence
- Network access in prose of documentationai-tools-setup/references/hermes.md:20Low confidence
- Network access in code of documentationai-tools-setup/references/hermes.md:25Low confidence
- Dangerous shell operation in code of documentationai-tools-setup/references/hermes.md:26Low confidence
- Network access in code of documentationai-tools-setup/references/hermes.md:26Low confidence
- Network access in code of documentationai-tools-setup/references/hermes.md:30Low confidence
- Network access in code of documentationai-tools-setup/references/hermes.md:31Low confidence
- Network access in prose of documentationai-tools-setup/references/openclaw.md:3Low confidence
- Network access in prose of documentationai-tools-setup/references/openclaw.md:4Low confidence
- Network access in prose of documentationai-tools-setup/references/openclaw.md:5Low confidence
- Network access in prose of documentationai-tools-setup/references/openclaw.md:26Low confidence
- Network access in prose of documentationai-tools-setup/references/openclaw.md:72Low confidence
- Network access in prose of documentationai-tools-setup/references/openclaw.md:74Low confidence
- Network access in prose of documentationai-tools-setup/references/openclaw.md:78Low confidence
- Network access in prose of documentationai-tools-setup/references/openclaw.md:79Low confidence
- Network access in prose of documentationai-tools-setup/references/openclaw.md:80Low confidence
- Dangerous shell operation in the frontmatter of the SKILL.mdplugins/claude/skills/computer-health-check/SKILL.md:16Low confidence
- Dangerous shell operation in the frontmatter of the SKILL.mdcomputer-health-check/SKILL.md:16Low confidence
- Dangerous shell operation in prose of documentationplugins/claude/skills/computer-health-check/references/linux.md:35Low confidence
- Dangerous shell operation in prose of documentationcomputer-health-check/references/linux.md:35Low confidence
- Dangerous shell operation in prose of documentationplugins/claude/skills/computer-health-check/references/linux.md:50Low confidence
- Dangerous shell operation in prose of documentationcomputer-health-check/references/linux.md:50Low confidence
- Dangerous shell operation in prose of documentationplugins/claude/skills/computer-health-check/references/macos.md:54Low confidence
- Dangerous shell operation in prose of documentationcomputer-health-check/references/macos.md:54Low confidence
- Network access in prose of the SKILL.mdplugins/claude/skills/desktop-commander-overview/SKILL.md:48Low confidence
- Network access in prose of the SKILL.mddesktop-commander-overview/SKILL.md:48Low confidence
- Dangerous shell operation in prose of the SKILL.mdplugins/claude/skills/terminal/SKILL.md:58Low confidence
- Dangerous shell operation in prose of the SKILL.mdterminal/SKILL.md:58Low confidence
- Network access in prose of the SKILL.mdplugins/claude/skills/terminal/SKILL.md:108Low confidence
- Network access in prose of the SKILL.mdterminal/SKILL.md:108Low confidence
- Network access in prose of the SKILL.mdplugins/claude/skills/terminal/SKILL.md:109Low confidence
- Network access in prose of the SKILL.mdterminal/SKILL.md:109Low confidence
- Network access in prose of the SKILL.mdplugins/claude/skills/terminal/SKILL.md:110Low confidence
- Network access in prose of the SKILL.mdterminal/SKILL.md:110Low confidence
- Network access in prose of the SKILL.mdplugins/claude/skills/terminal/SKILL.md:111Low confidence
- Network access in prose of the SKILL.mdterminal/SKILL.md:111Low confidence
- Dangerous shell operation in prose of the SKILL.mdplugins/claude/skills/terminal/SKILL.md:112Low confidence
- Dangerous shell operation in prose of the SKILL.mdterminal/SKILL.md:112Low confidence
- Network access in prose of the SKILL.mdplugins/claude/skills/terminal/SKILL.md:112Low confidence
- Network access in prose of the SKILL.mdterminal/SKILL.md:112Low confidence
and 92 more
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Native | Manifest .claude-plugin/marketplace.json | .claude/skills/<skill-name>/ |
| Codex | Compatible | Standard SKILL.md plugins/claude/skills/ai-tools-setup/SKILL.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md plugins/claude/skills/ai-tools-setup/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Native | Manifest gemini-extension.json | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md plugins/claude/skills/ai-tools-setup/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md plugins/claude/skills/ai-tools-setup/SKILL.md | .opencode/skills/<skill-name>/ |
- Native The repository ships a manifest, folder or frontmatter field for this agent.
- Compatible A standard SKILL.md that this agent can read.
- Unknown No evidence either way.
Derived automatically from paths, manifests and frontmatter. Not tested on each agent.
Explore alternatives and comparisons
Add this badge to your README
It updates every night and links to this page.
[](https://hotskills.dev/skill/wonderwhy-er-desktopcommandermcp/)