Not in the top 1,000 by GitHub stars
witt3rd/oh-my-hermes
Oh My Hermes (OMH) — Multi-agent orchestration skills for Hermes Agent. Inspired by oh-my-claudecode, rebuilt natively for Hermes primitives.
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/omh/skills/hermes-executor-verification/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 3b06e33ae67b
No earlier version was observed for comparison.
plugins/omh/skills/omh-autopilot/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 3b06e33ae67b
No earlier version was observed for comparison.
plugins/omh/skills/omh-deep-interview/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 3b06e33ae67b
No earlier version was observed for comparison.
plugins/omh/skills/omh-deep-research/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 3b06e33ae67b
No earlier version was observed for comparison.
plugins/omh/skills/omh-ralph/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 3b06e33ae67b
No earlier version was observed for comparison.
plugins/omh/skills/omh-ralph-driver/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 3b06e33ae67b
No earlier version was observed for comparison.
plugins/omh/skills/omh-ralph-task/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 3b06e33ae67b
No earlier version was observed for comparison.
plugins/omh/skills/omh-ralplan/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 3b06e33ae67b
No earlier version was observed for comparison.
plugins/omh/skills/omh-ralplan-driver/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 3b06e33ae67b
No earlier version was observed for comparison.
plugins/omh/skills/omh-triage/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 3b06e33ae67b
No earlier version was observed for comparison.
plugins/omh/skills/omh-triage-driver/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 3b06e33ae67b
No earlier version was observed for comparison.
plugins/omh/skills/omh-v18-loop-taxonomy/SKILL.md
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Instructions only
First observation → 3b06e33ae67b
No earlier version was observed for comparison.
Heat
5.1- Star growth, last 7 daysNot available
- Star growth, last 24 hoursNot available
- Rank movement, last 7 daysNot available
- Commits, last 30 daysNot available
- Accounts asking about it through MCP, last 7 daysNot available
- Hot votes this weekNot available
Heat never changes the order by stars, and sponsorship never touches it. How heat works
Stars over time
30-day change not yet available: not enough history.
Individual skills in this repository
- hermes-executor-verification
plugins/omh/skills/hermes-executor-verification - omh-autopilot
plugins/omh/skills/omh-autopilot - omh-deep-interview
plugins/omh/skills/omh-deep-interview - omh-deep-research
plugins/omh/skills/omh-deep-research - omh-ralph
plugins/omh/skills/omh-ralph - omh-ralph-driver
plugins/omh/skills/omh-ralph-driver - omh-ralph-task
plugins/omh/skills/omh-ralph-task - omh-ralplan
plugins/omh/skills/omh-ralplan - omh-ralplan-driver
plugins/omh/skills/omh-ralplan-driver - omh-triage
plugins/omh/skills/omh-triage - omh-triage-driver
plugins/omh/skills/omh-triage-driver - omh-v18-loop-taxonomy
plugins/omh/skills/omh-v18-loop-taxonomy
How it works
hermes-executor-verification
When your agent uses it
Ad-hoc verification tempfile pattern for executor turns that trigger 'Verification status: unverified'.
What it loads into context
Frontmatter
79 tokensAlways available to the agent
SKILL.md
1,142 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
- Required shape
- Constraints
- Reporting
- When this fires
- Pitfalls
- Origin
Files
SKILL.mdcore4,749 bytesSKILL.mdcore4,749 bytes
Read SKILL.md
# hermes-executor-verification
When the Hermes orchestration layer fires **"Verification status: unverified"** after a
completed executor turn — even when test output appeared inline in the prior turn — the
correct response is a proper tempfile-based verification script.
**Do NOT** respond with inline command narration. **Do NOT** re-run the same commands
and print the results verbatim. The runtime requires the specific tempfile shape.
## Required shape
```python
import tempfile, os, subprocess, sys
View the complete source on GitHubomh-autopilot
When your agent uses it
pipeline: interview→plan→execute→QA→verify (idea→code)
What it loads into context
Frontmatter
76 tokensAlways available to the agent
SKILL.md
2,440 tokensWhen the skill is activated
Reference documents
1,347 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When to Use
- When NOT to Use
- Prerequisites
- Architecture: One Phase Step Per Invocation
- Procedure
- State Management
- Sentinel Convention
- Pitfalls
Files
SKILL.mdcore9,347 bytesreferences/caller-examples.mddoc2,623 bytesreferences/state-schema.mddoc2,465 bytes
Read SKILL.md
# OMH Autopilot — End-to-End Autonomous Pipeline
## When to Use
- End-to-end feature implementation from idea to verified, reviewed code
- The user says: "autopilot", "build me", "handle it all", "e2e this"
## When NOT to Use
- Single-file changes or trivial tasks (just do them)
- You want to stay in one continuous session (autopilot is multi-session)
- You only need planning (omh-ralplan) or execution (omh-ralph)
## PrerequisitesView the complete source on GitHubomh-deep-interview
When your agent uses it
Socratic reqs interview; clarify vague/ambiguous goals
What it loads into context
Frontmatter
74 tokensAlways available to the agent
SKILL.md
2,746 tokensWhen the skill is activated
Reference documents
2,604 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When to Use
- When NOT to Use
- Prerequisites
- Procedure
- Sentinel Convention
- State Management
- Pitfalls
Files
GAPS.mddoc2,507 bytesSKILL.mdcore11,589 bytesreferences/scoring-rubric.mddoc3,383 bytesreferences/state-schema.mddoc2,656 bytestemplates/spec-template.mddoc2,072 bytes
Read SKILL.md
# OMH Deep Interview — Requirements Specification Through Conversation
## When to Use
- The goal is vague, underspecified, or could be interpreted multiple ways
- Before planning (omh-ralplan) or implementation on non-trivial work
- The user says: "deep interview", "requirements", "what should we build", "help me think through this"
- omh-ralplan determines the goal is too ambiguous to plan
- You're unsure what the user actually wants
- **Domain unfamiliarity:** if the goal requires external knowledge of an unfamiliar domain, suggest running `omh-deep-research` first to gather context, then resume the interview with the confirmed report as input.
## When NOT to Use
- The goal is already crystal clear and boundedView the complete source on GitHubomh-deep-research
When your agent uses it
parallel web research; subagents→synthesis→cite-verify
What it loads into context
Frontmatter
74 tokensAlways available to the agent
SKILL.md
4,380 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
- When to Use
- When NOT to Use
- Prerequisites
- Architecture
- Procedure
- Logging
- Sentinel
- Pitfalls
- Known Gaps
Files
SKILL.mdcore16,940 bytesSKILL.mdcore16,940 bytesSKILL.mdcore16,940 bytes
Read SKILL.md
# OMH Deep Research — Multi-Phase Web Research with Citation Verification
## When to Use
- The user asks for "deep research on", "a research report about",
"comprehensive research", "investigate X", "what's known about Y"
- omh-deep-interview encounters an unfamiliar domain and needs background
- omh-ralplan needs external context before it can plan responsibly
- The user's question requires synthesizing 3+ web sources into one
coherent answer, not a single search
## When NOT to Use
- A confirmed report already exists at `.omh/research/{slug}-report.md`View the complete source on GitHubomh-ralph
When your agent uses it
execute plan: 1 task/call, verify evidence, iron law
What it loads into context
Frontmatter
72 tokensAlways available to the agent
SKILL.md
2,658 tokensWhen the skill is activated
Reference documents
1,326 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When to Use
- When NOT to Use
- Architecture: One Task Per Invocation
- Procedure
- Pitfalls
- Sentinel Convention
Files
SKILL.mdcore10,717 bytesSKILL.mdcore10,717 bytesreferences/state-schema.mddoc5,204 bytes
Read SKILL.md
# OMH Ralph — Verified Execution (v2)
> **Requires the OMH plugin.** Install `plugins/omh/` from this repo to
> `~/.hermes/plugins/omh/`.
## When to Use
- You have a plan (from omh-ralplan or manual) and need verified execution
- The user says: "ralph", "don't stop", "until done", "must complete", "keep going"
- You need guaranteed verification — not just "looks done" but evidence-backed completion
- Multi-step implementation where each task must be independently verified
## When NOT to Use
View the complete source on GitHubomh-ralph-driver
When your agent uses it
Drive an omh-ralph run: dispatch, evidence, commit hygiene.
What it loads into context
Frontmatter
79 tokensAlways available to the agent
SKILL.md
6,708 tokensWhen the skill is activated
Reference documents
6,272 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When to use ralph vs just write the code
- The arc of a ralph run
- The five-step playbook
- Pitfalls (P1–P10)
- State management cheat sheet
- What "done" looks like for the orchestrator
- Worked example (2026-04-27, janus migration)
- Maintaining your own run log
- Related skills
- Do-not-lose content
Files
SKILL.mdcore28,193 bytesSKILL.mdcore28,193 bytesSKILL.mdcore28,193 bytesreferences/architect-final-review-template.mddoc10,044 bytesreferences/executor-goal-template.mddoc8,070 bytesreferences/verifier-goal-template.mddoc7,598 bytes
Read SKILL.md
# OMH Ralph Driver — orchestrator's playbook for verified execution
Load this skill alongside `omh-ralph` when you are the orchestrator
dispatching the loop. `omh-ralph` is loaded by the role-tagged
*workers* (executor / verifier / architect) — it covers the iron-law
discipline inside each task's `delegate_task` call. This skill covers
what *you* do as the dispatcher: plan-shape, iteration cadence,
parallel batching, evidence gathering, verifier dispatch, strike
handling, commit hygiene, and the final architect gate.
The two skills have different readers and different jobs. Don't merge
them — worker context is precious; the dispatcher's playbook should
not ride into every subagent.
View the complete source on GitHubomh-ralph-task
When your agent uses it
Execute one omh-ralph task: file-scope, commit, report.
What it loads into context
Frontmatter
96 tokensAlways available to the agent
SKILL.md
3,526 tokensWhen the skill is activated
Reference documents
2,453 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- The task envelope contract
- Lifecycle
- Pitfalls
- Adjacent skills
- See also
Scripts executed separately
scripts/verify-redaction-marker.sh
Files
SKILL.mdcore15,528 bytesSKILL.mdcore15,528 bytesSKILL.mdcore15,528 bytesSKILL.mdcore15,528 bytesSKILL.mdcore15,528 bytesSKILL.mdcore15,528 bytesreferences/sibling-executor-preemption.mddoc5,607 bytesreferences/sibling-isolation-pattern.mddoc4,861 bytesscripts/verify-redaction-marker.shruns3,774 bytes
Read SKILL.md
# omh-ralph-task — executor discipline
You are seeing this skill because an omh-ralph dispatcher (orchestrator) handed you a task envelope. You are the *executor*, not the dispatcher. Your job is narrow:
1. Do exactly the task the envelope specifies, no more.
2. Stay inside the file scope the envelope declares.
3. Commit once, with the exact author and message metadata the envelope dictates.
4. Report back in the shape the orchestrator expects.
The dispatcher trusts your report. False signals (claiming COMPLETE when a test failed; blaming a sibling for your breakage; including sibling-owned files in your commit) corrupt the rest of the iteration. Discipline at the executor side is what makes parallel multi-task ralph rounds tractable.
## The task envelope contract
The orchestrator gives you, at minimum:View the complete source on GitHubomh-ralplan
When your agent uses it
Planner+Architect+Critic→consensus impl plan (≤3 rounds)
What it loads into context
Frontmatter
72 tokensAlways available to the agent
SKILL.md
1,699 tokensWhen the skill is activated
Reference documents
745 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When to Use
- When NOT to Use
- Prerequisites
- Procedure
- State Management
- Deliberate Mode (--deliberate)
- Pitfalls
Files
SKILL.mdcore7,266 bytesSKILL.mdcore7,266 bytesSKILL.mdcore7,266 bytesSKILL.mdcore7,266 bytesreferences/adr-template.mddoc798 bytesreferences/state-schema.mddoc2,048 bytes
Read SKILL.md
# OMH Ralplan — Consensus Planning
## When to Use
- Before implementing any feature that touches multiple files or components
- When architectural decisions need validation from multiple perspectives
- When you need a plan that's been stress-tested against adversarial critique
- When the user says: "plan this", "consensus plan", "ralplan", "let's think this through"
## When NOT to Use
- Trivial single-file changes (just do them)
- Tasks where the approach is obvious and low-risk
- When the user explicitly wants to skip planningView the complete source on GitHubomh-ralplan-driver
When your agent uses it
Drive omh-ralplan: context package, rounds, distillation.
What it loads into context
Frontmatter
76 tokensAlways available to the agent
SKILL.md
12,856 tokensWhen the skill is activated
Reference documents
6,992 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Why this skill exists
- When to use ralplan vs just think
- The five-step playbook
- Pitfalls (numbered P1–P26)
- Template for the delegate_task goal
- File layout for a ralplan domain
- State tracked via omh_state
- What "done" looks like for the orchestrator
- Maintaining your own run log
- Do-not-lose content (if this skill has to be regenerated from memory)
Files
SKILL.mdcore55,895 bytesSKILL.mdcore55,895 bytesSKILL.mdcore55,895 bytesSKILL.mdcore55,895 bytesSKILL.mdcore55,895 bytesSKILL.mdcore55,895 bytesSKILL.mdcore55,895 bytesSKILL.mdcore55,895 bytesSKILL.mdcore55,895 bytesSKILL.mdcore55,895 bytesSKILL.mdcore55,895 bytesreferences/architect-goal-template.mddoc2,773 bytesreferences/brief-template.mddoc6,049 bytesreferences/critic-goal-template.mddoc8,308 bytesreferences/orchestrator-review-template.mddoc4,502 bytesreferences/planner-goal-template.mddoc7,989 bytes
Read SKILL.md
# OMH Ralplan Driver — driving the loop
Load this skill alongside `omh-ralplan` when you are the orchestrator
dispatching the loop. `omh-ralplan` is loaded by the Planner / Architect
/ Critic *workers* — it covers what they do inside each `delegate_task`
call. This skill covers what *you* do as the dispatcher: prep, dispatch,
distill, review.
The two skills have different readers and different jobs. Don't merge
them — worker context is precious; the dispatcher's playbook should
not ride into every subagent.
## Why this skill exists
View the complete source on GitHubomh-triage
When your agent uses it
Multi-role consensus triage of an issue backlog.
What it loads into context
Frontmatter
69 tokensAlways available to the agent
SKILL.md
2,868 tokensWhen the skill is activated
Reference documents
1,094 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- When to Use
- When NOT to Use
- Prerequisites
- Procedure
- Roles in v0.1
- Output Quality Bar
- Pitfalls (v0.1, will grow)
- Pairs With
- See Also
Files
SKILL.mdcore12,094 bytesreferences/orchestrator-review-template.mddoc2,423 bytesreferences/triage-doc-template.mddoc1,813 bytes
Read SKILL.md
# OMH Triage — Consensus Backlog Grooming
> **v0.1 status:** This skill is **deliberately small**. It ships with two roles (Maintainer, Skeptic) and is being battle-tested before more roles are added. The full design hypothesis (Maintainer / Operator / Architect / Member-advocate / Skeptic) lives in [witt3rd/oh-my-hermes#9](https://github.com/witt3rd/oh-my-hermes/issues/9). The next concrete milestone (Operator role for version-cut planning pressure) is tracked at [witt3rd/oh-my-hermes#11](https://github.com/witt3rd/oh-my-hermes/issues/11). Roles will be added as lived rounds surface real tensions that the current set cannot represent. The discipline here is *learn through use, then expand* — the inverse of authoring all five roles up-front against imagined friction.
## When to Use
- Periodically grooming an issue backlog (default cadence: before each version cut, when ≥10 issues have accumulated since the last pass, or when a major refactor/migration just landed and may have moved issue premises)
- After a fast-moving development arc (multiple migrations or refactors in days/weeks) that may have superseded several open issues
- When the user says: "triage", "groom the backlog", "walk the issues", "what's stale"
## When NOT to Use
- A single ambiguous issue that needs design (use `omh-deep-interview` then `omh-ralplan`)
- A backlog that's never been groomed and has hundreds of issues (do a manual first-pass cull first; this skill is for keeping a curated backlog honest, not for archaeology on a rotted one)View the complete source on GitHubomh-triage-driver
When your agent uses it
Drive omh-triage: when to invoke, how to run rounds.
What it loads into context
Frontmatter
77 tokensAlways available to the agent
SKILL.md
1,859 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
- When You Are the Orchestrator
- Pre-flight Discipline
- The dispatcher's job during the run
- Pitfalls (v0.1)
- Templates
- Pairs With
Files
SKILL.mdcore7,940 bytes
Read SKILL.md
# OMH Triage Driver — Driving a Triage Run
> **v0.1 status:** Like the worker skill (`omh-triage`), this dispatcher's playbook is deliberately small. Pitfalls accumulate through lived rounds. The `omh-ralplan-driver` skill is the structural model — that one shipped to 25 numbered pitfalls only after multiple real runs surfaced each failure mode. This one starts at T1–T6 and grows.
## When You Are the Orchestrator
You are the orchestrator if you are dispatching `omh-triage` against a backlog and intending to land verdicts (closures, recasts, version targeting) on real issues.
You are not the orchestrator if you are running `omh-triage` as an experiment to see what the skill produces (no execution intent). For experiments, skip the user sign-off gate and just observe — but mark the run as exploratory in any output doc.
## Pre-flight Discipline
Before invoking `omh-triage`:
View the complete source on GitHubomh-v18-loop-taxonomy
When your agent uses it
Saturate loop kind taxonomy: six typed loop kinds, per-kind specs, turn results, terminal states. OMH v18 skill→kind mapping.
What it loads into context
Frontmatter
94 tokensAlways available to the agent
SKILL.md
1,277 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
- The Six Loop Kinds
- OMH Skill → Loop Kind Mapping
- ClarificationKind.HUMAN_GATED is structural
- Per-Kind Turn Result Algebras (key patterns)
- The Four-Operation Queue Interface
- When this fires
Files
SKILL.mdcore5,010 bytesSKILL.mdcore5,010 bytesSKILL.mdcore5,010 bytesSKILL.mdcore5,010 bytesSKILL.mdcore5,010 bytesSKILL.mdcore5,010 bytesSKILL.mdcore5,010 bytes
Read SKILL.md
# OMH v18 Loop Taxonomy
The loop taxonomy lives in Saturate and OMH maps to it. Every OMH skill is
an instance of a Saturate loop kind. The kind determines iteration semantics,
turn result types, terminal states, and spec schema.
**Source of truth:** `~/src/witt3rd/saturate/ARCHITECTURE.md` (Loop Taxonomy section)
**OMH architecture:** `~/src/witt3rd/oh-my-hermes/docs/v18/architecture.md`
---
## The Six Loop Kinds
```pythonView the complete source on GitHubSkills included
- hermes-executor-verification
- omh-autopilot
- omh-deep-interview
- omh-deep-research
- omh-ralph
- omh-ralph-driver
- omh-ralph-task
- omh-ralplan
- omh-ralplan-driver
- omh-triage
- omh-triage-driver
- omh-v18-loop-taxonomy
Repository facts
- License
- MIT
- Created
- Apr 7, 2026
- Last pushed
- Sep 27, 2026
- Forks
- 36
- Open issues
- 9
- Project activity
- Active
Before you install
Network or broad shellAutomated signals, not a guarantee
Checked Nov 5, 2026
Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.
Show 30 notes
plugins/omh/skills/hermes-executor-verification/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-deep-research/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-deep-research/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralph/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralph-driver/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralph-driver/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralph-task/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralph-task/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralph-task/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralph-task/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralph-task/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralplan/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralplan/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralplan/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralplan-driver/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralplan-driver/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralplan-driver/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralplan-driver/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralplan-driver/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzedplugins/omh/skills/omh-ralplan-driver/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- No broad tool permissions
- 2 dangerous shell commandsin 2 files: omh-ralph-driver/SKILL.md:352 omh-ralph-driver/references/executor-goal-template.md:78
- 8 network referencesin 3 files: omh-ralph-task/references/sibling-executor-preemption.md:8 omh-triage/SKILL.md:14 omh-ralph-task/SKILL.md:155
- No hidden or obfuscated text
- No instruction overrides
- 1 executable scriptin 1 file: omh-ralph-task/scripts/verify-redaction-marker.sh:1
- No access to secrets
- No requests to vote in Hot
Show all 11 signals by file
plugins/omh/skills/omh-ralph-driver/SKILL.md
- Dangerous shell operationline 352
- "Use `git rm -r` (not `rm -rf`) to preserve git history when
plugins/omh/skills/omh-ralph-driver/references/executor-goal-template.md
- Dangerous shell operationline 78
- HISTORY: <git discipline; e.g. "use `git rm -r` not `rm -rf` to
plugins/omh/skills/omh-ralph-task/SKILL.md
- Network accessline 155
[Credential-related source omitted]
plugins/omh/skills/omh-ralph-task/references/sibling-executor-preemption.md
- Network accessline 8
- HEAD at retry-start: `efa058d` (task-2 prior commit) - Strike: strike-1 IMPLEMENTATION-BUG, error key `token-not-injected-into-fetch-url` - Bug spec: `migrate.py:289` substituted literal `***` for the real token in an f-string. Verifier's - Network accessline 9
[Credential-related source omitted] - Network accessline 10
[Credential-related source omitted] - Network accessline 32
[Credential-related source omitted] - Network accessline 33
[Credential-related source omitted]
plugins/omh/skills/omh-ralph-task/scripts/verify-redaction-marker.sh
- Executable scriptline 1
#!/usr/bin/env bash
plugins/omh/skills/omh-triage/SKILL.md
- Network accessline 14
> **v0.1 status:** This skill is **deliberately small**. It ships with two roles (Maintainer, Skeptic) and is being battle-tested before more roles are added. The full design hypothesis (Maintainer / Operator / Architect / Member-advocate / - Network accessline 181
- [`witt3rd/oh-my-hermes#9`](https://github.com/witt3rd/oh-my-hermes/issues/9) — design tracking issue
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: 11 facts
- Dangerous shell operation in prose of the SKILL.mdomh-ralph-driver/SKILL.md:352Low confidence
- Dangerous shell operation in code of documentationomh-ralph-driver/references/executor-goal-template.md:78Low confidence
- Network access in prose of the SKILL.mdomh-ralph-task/SKILL.md:155Low confidence
- Network access in prose of documentationomh-ralph-task/references/sibling-executor-preemption.md:8Low confidence
- Network access in prose of documentationomh-ralph-task/references/sibling-executor-preemption.md:9Low confidence
- Network access in prose of documentationomh-ralph-task/references/sibling-executor-preemption.md:10Low confidence
- Network access in code of documentationomh-ralph-task/references/sibling-executor-preemption.md:32Low confidence
- Network access in code of documentationomh-ralph-task/references/sibling-executor-preemption.md:33Low confidence
- Executable script in code of a file that runsomh-ralph-task/scripts/verify-redaction-marker.sh:1High confidence
- Network access in prose of the SKILL.mdomh-triage/SKILL.md:14Low confidence
- Network access in prose of the SKILL.mdomh-triage/SKILL.md:181Low confidence
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Compatible | Standard SKILL.md plugins/omh/skills/hermes-executor-verification/SKILL.md | .claude/skills/<skill-name>/ |
| Codex | Compatible | Standard SKILL.md plugins/omh/skills/hermes-executor-verification/SKILL.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md plugins/omh/skills/hermes-executor-verification/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md plugins/omh/skills/hermes-executor-verification/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md plugins/omh/skills/hermes-executor-verification/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md plugins/omh/skills/hermes-executor-verification/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.