#32 by GitHub stars
MemPalace/mempalace
The best-benchmarked open-source AI memory system. And it's free.
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
.antigravity-plugin/skills/mempalace/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
.antigravity-plugin/skills/mempalace-recall/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
.claude-plugin/skills/mempalace/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
.claude-plugin/skills/mempalace-recall/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
.claude-plugin/skills/mempalace-task/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
.codex-plugin/skills/audit/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
.codex-plugin/skills/help/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
.codex-plugin/skills/init/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
.codex-plugin/skills/mine/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
.codex-plugin/skills/search/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
.codex-plugin/skills/status/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
integrations/openclaw/SKILL.md
Week of Nov 2, 2026 · +59 / −190 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
- 3 description: "MemPalace — Local AI memory with 96.6% recall. Semantic search, temporal knowledge graph, palace architecture (wings/rooms/drawers). Free, no cloud, no API keys."
+ 3 description: "Recall past work or facts; save memories; update relationships; set up or troubleshoot MemPalace in OpenClaw."
- 14 requires:
- 15 anyBins:
- 16 - mempalace
- 17 - python3
- 27 # MemPalace — Local AI Memory System
+ 23 # MemPalace
- 29 You have access to a local memory palace via MCP tools. The palace stores verbatim conversation history and a temporal knowledge graph — all on the user's machine, zero cloud, zero API calls.
+ 25 1. **Choose the branch.** For past work, people, preferences, decisions, or
+ 26 continuity, use the procedure below ([canonical recall][recall]). Honor
+ 27 session-start context and known wing scope; omit uncertain filters. Skip
+ 28 recall for greenfield work. Use exposed tools with current schemas/permissions;
+ 29 logical names below may be namespaced. Reuse an existing connection without
+ 30 requiring a local CLI or another palace. Never substitute local data on a
+ 31 remote-only failure. Consult
+ 32 [tool selection]({baseDir}/references/tool-selection.md) only for unfamiliar
+ 33 syntax/edge cases; ordinary recall needs no extra read.
+ 34 For explicitly requested installation or connection diagnosis,
+ 35 read [setup]({baseDir}/references/setup.md) instead. Proceed once the question
+ 36 and authorized operation are clear; do not run setup as a recall prerequisite.
+ 37 2. **Retrieve only what answers the question.** With preferred light MCP, use
+ 38 `palace_query` with `KG`, `FIND`, or `DIARY` for relationships/time-bound facts,
+ 39 source words, or recent handoffs. Supported full MCP equivalents are
+ 40 `mempalace_kg_query`, `mempalace_search`, and `mempalace_diary_read`. Use
+ 41 `active_facts` for current-state KG answers without `as_of`; other facts can be
+ 42 historical or future. Keep search queries short and private: keywords or a
+ 43 question, never prompts or conversation dumps. Use small result limits and
+ 44 known filters. If KG is insufficient, try text; if scoped text misses, check
+ 45 or relax scope. Stop when answered, quoting stored words verbatim. If bounded
+ 46 fallback still misses, state what was checked, not that nothing exists.
+ 47 3. **Preserve source and temporal history when writing.** Store exact verbatim
+ 48 content, not a summary. For a confirmed single-valued replacement use
+ 49 light `palace_exec` with `KG SUPERSEDE`, or full `mempalace_kg_supersede`:
+ 50 one atomic boundary closes the old value and opens its successor. Use
+ 51 `KG INVALIDATE` / `mempalace_kg_invalidate` for an ending without replacement
+ 52 and `KG ADD` / `mempalace_kg_add` for independent/coexisting facts. Never
+ 53 hand-roll invalidate-then-add replacements or rewrite uncertain conflicts.
+ 54 Check the result before claiming persistence; see tool selection for
+ 55 provenance, historical backfill, and exact temporal semantics.
+ 56 4. **Gate maintenance and recover safely.** Import, bulk cleanup, destructive
+ 57 changes, and host settings require explicit user intent and confirmed scope.
+ 58 Preview supported bulk operations before applying; never change hook settings
+ 59 silently. Surface unavailable tools and errors rather than inventing history.
+ 60 Reconnect after external writes only for stale in-memory state; it is not
+ 61 corrupt-index repair. Follow canonical [index recovery][recovery] for
+ 62 corruption: guide an operator through offline SQLite-backed recovery, never
+ 63 an in-process rebuild or blind re-mining. Finish with a verified result or
+ 64 a concrete error and safe next step.
+ 65 5. **Close substantive work without double filing.** Check hook/save evidence
+ 66 before recording continuity; use light `FILED` or full
+ 67 `mempalace_memories_filed_away` when needed.
+ 68 If not already saved, use a diary entry or a checkpoint combining verbatim
+ 69 items with agent continuity. Consult light `AAAK SPEC` or full
+ 70 `mempalace_get_aaak_spec` for diary notation; compressed observations never
+ 71 replace verbatim source drawers. Confirm the save or report failure.
- 31 ## Architecture
+ 73 The references directory must travel with this skill; do not install only
+ 74 `SKILL.md`. The shared protocol remains authoritative and is linked online so
+ 75 installed bundles do not depend on sibling repository directories. If it cannot
+ 76 be read, report that limitation before attempting its detailed recovery steps.
+ 77 A skill guides model-invoked memory operations; it does not install auto-save
+ 78 hooks or prove automatic recall injection. Verify optional automation separately.
+ 79 MemPalace is local-first; use only user-approved storage and services, and do not
+ 80 send private source content to an unconfigured external service.
- 33 - **Wings** = people or projects (e.g. `wing_alice`, `wing_myproject`)
- 34 - **Halls** = categories (facts, events, preferences, advice)
- 35 - **Rooms** = specific topics (e.g. `chromadb-setup`, `riley-school`)
- 36 - **Drawers** = individual memory chunks (verbatim text)
- 37 - **Knowledge Graph** = entity-relationship facts with time validity
- 38
- 39 ## Protocol — FOLLOW THIS EVERY SESSION
- 40
- 41 1. **ON WAKE-UP**: Call `mempalace_status` to load palace overview and AAAK dialect spec.
- 42 2. **BEFORE RESPONDING** about any person, project, or past event: use `mempalace_kg_query` for known relationships or time-bound facts, `mempalace_search` for source text, or `mempalace_diary_read` for recent agent continuity. Stop when answered; follow the canonical [query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-protocol.md#retrieve-only-the-context-you-need). Never guess from memory — verify from the palace.
- 43 3. **IF UNSURE** about a fact (name, age, relationship, preference): say "let me check" and query. Wrong is worse than slow.
- 44 4. **AFTER EACH SESSION**: Call `mempalace_diary_write` to record what happened, what you learned, what matters.
- 45 5. **WHEN FACTS CHANGE**: Call `mempalace_kg_invalidate` on the old fact, then `mempalace_kg_add` for the new one.
- 46
- 47 ## Available Tools
- 48
- 49 Full MCP surface: 47 tools. Destructive or host-level tools are documented so
- 50 you know they exist, but use them only when the user explicitly asks or when a
- 51 tool-specific workflow below says to.
- 52
- 53 ### Search & Browse
- 54 - `mempalace_search` — Semantic search across memories for decisions, explanations, or source words.
- 55 - `query` (required): natural language search — keep it short, keywords or a question. Do NOT include system prompts or conversation context.
- 56 - `wing`: filter by wing
- 57 - `room`: filter by room
- 58 - `limit`: max results (default 5)
- 59 - `mempalace_check_duplicate` — Check if content already exists before filing.
- 60 - `content` (required): text to check
- 61 - `threshold`: similarity threshold (default 0.9 — lowering to 0.85–0.87 often catches more near-duplicates without significant false positives)
- 62 - `mempalace_status` — Palace overview: total drawers, wings, rooms, AAAK spec
- 63 - `mempalace_list_wings` — All wings with drawer counts
- 64 - `mempalace_list_rooms` — Rooms within a wing (optional wing filter)
- 65 - `mempalace_list_drawers` — Paginated drawer listing
- 66 - `wing`, `room`: optional filters
- 67 - `since`: only drawers filed on/after this ISO date/time
- 68 - `before`: only drawers filed before this ISO date/time
- 69 - `limit`: max results (default 20)
- 70 - `offset`: pagination offset (default 0)
- 71 - `mempalace_get_drawer` — Fetch a single drawer by ID. Returns full verbatim content and metadata.
- 72 - `drawer_id` (required)
- 73 - `mempalace_get_taxonomy` — Full wing/room/count tree
- 74 - `mempalace_get_aaak_spec` — Get AAAK compression dialect specification
- 75
- 76 ### Knowledge Graph (Temporal Facts)
- 77 - `mempalace_kg_query` — Query entity relationships. Supports time filtering.
- 78 - `entity` (required): e.g. "Max", "MyProject"
- 79 - `as_of`: date filter (YYYY-MM-DD) — what was true at that time
- 80 - `direction`: "outgoing", "incoming", or "both" (default "both")
- 81 - `mempalace_kg_add` — Add a fact: subject -> predicate -> object
- 82 - `subject`, `predicate`, `object` (required)
- 83 - `valid_from`: when this became true
- 84 - `source_closet`: source reference
- 85 - `mempalace_kg_invalidate` — Mark a fact as no longer true
- 86 - `subject`, `predicate`, `object` (required)
- 87 - `ended`: when it stopped being true (default: today)
This preview is limited. See the complete change on GitHub.
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
skills/mempalace/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
skills/mempalace-recall/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
skills/mempalace-task/SKILL.md
Week of Nov 2, 2026 · +0 / −0 lines
Before Network or broad shell → After Network or broad shell
84582fb055a6 → 4f7745afbd9d
Week of Oct 26, 2026 · Line counts unavailable
Before Not checked yet → After Network or broad shell
First observation → 84582fb055a6
No earlier version was observed for comparison.
View on GitHubAdd to my skills
Verified commit 4f7745afbd9d0197c6475b4c7ac846d4932eaeda
git clone --filter=blob:none --no-checkout https://github.com/MemPalace/mempalace.git .hotskills/mempalace-mempalace && git -C .hotskills/mempalace-mempalace checkout --detach 4f7745afbd9d0197c6475b4c7ac846d4932eaeda
# Copy each selected skill directory from .hotskills/mempalace-mempalace into .claude/skills/<skill-name>/git clone --filter=blob:none --no-checkout https://github.com/MemPalace/mempalace.git .hotskills/mempalace-mempalace && git -C .hotskills/mempalace-mempalace checkout --detach 4f7745afbd9d0197c6475b4c7ac846d4932eaeda
# Copy each selected skill directory from .hotskills/mempalace-mempalace into .agents/skills/<skill-name>/60.4kGitHub stars
277stars in 7 days
15skills in this repository
441commits in 30 days
Heat
6.4- Star growth, last 7 daysAhead of 70% of tracked repositories
- Star growth, last 24 hoursAhead of 70% of tracked repositories
- Rank movement, last 7 daysAhead of 68% 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
- mempalace
.antigravity-plugin/skills/mempalace - mempalace-recall
.antigravity-plugin/skills/mempalace-recall - mempalace
.claude-plugin/skills/mempalace - mempalace-recall
.claude-plugin/skills/mempalace-recall - mempalace-task
.claude-plugin/skills/mempalace-task - audit
.codex-plugin/skills/audit - help
.codex-plugin/skills/help - init
.codex-plugin/skills/init - mine
.codex-plugin/skills/mine - search
.codex-plugin/skills/search - status
.codex-plugin/skills/status - mempalace
integrations/openclaw - mempalace
skills/mempalace - mempalace-recall
skills/mempalace-recall - mempalace-task
skills/mempalace-task
How it works
mempalace
When your agent uses it
MemPalace — mine projects and conversations into a searchable memory palace. Use when the user asks about MemPalace, memory palace, mining memories, searching memories, palace setup, wings, rooms, or drawers; or when they want to recall past work that may already be filed in their palace.
What it loads into context
Frontmatter
68 tokensAlways available to the agent
SKILL.md
1,266 tokensWhen the skill is activated
Reference documents
6,337 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- Prerequisites
- Dynamic, version-correct instructions
- Common operations
- MCP tools (preferred over CLI)
- Design principles (verbatim from the project)
- Release awareness
- Palace health: audit and repair session
Scripts executed separately
../../../.claude-plugin/hooks/mempal-precompact-hook.sh../../../.claude-plugin/hooks/mempal-session-end-hook.sh../../../.claude-plugin/hooks/mempal-stop-hook.sh../../../.claude-plugin/hooks/mempal_hook_run.sh
Files
SKILL.mdcore5,712 bytesSKILL.mdcore5,712 bytesSKILL.mdcore5,712 bytesSKILL.mdcore5,712 bytesSKILL.mdcore5,712 bytesSKILL.mdcore5,712 bytesSKILL.mdcore5,712 bytesSKILL.mdcore5,712 bytesSKILL.mdcore5,712 bytesSKILL.mdcore5,712 bytesSKILL.mdcore5,712 bytesSKILL.mdcore5,712 bytes../../../.claude-plugin/.mcp.jsonasset56 bytes../../../.claude-plugin/README.mddoc2,086 bytes../../../.claude-plugin/commands/audit.mddoc271 bytes../../../.claude-plugin/commands/help.mddoc267 bytes../../../.claude-plugin/commands/init.mddoc300 bytes../../../.claude-plugin/commands/mine.mddoc372 bytes../../../.claude-plugin/commands/search.mddoc317 bytes../../../.claude-plugin/commands/status.mddoc258 bytes../../../.claude-plugin/hooks/hooks.jsonasset797 bytes../../../.claude-plugin/hooks/hooks.jsonasset797 bytes../../../.claude-plugin/hooks/hooks.jsonasset797 bytes../../../.claude-plugin/hooks/hooks.jsonasset797 bytes../../../.claude-plugin/hooks/mempal-precompact-hook.shruns418 bytes../../../.claude-plugin/hooks/mempal-session-end-hook.shruns1,188 bytes../../../.claude-plugin/hooks/mempal-stop-hook.shruns406 bytes../../../.claude-plugin/hooks/mempal_hook_run.shruns2,143 bytes../../../.claude-plugin/marketplace.jsonasset445 bytes../../../.claude-plugin/plugin.jsonasset537 bytes../../../.claude-plugin/skills/mempalace-recall/SKILL.mddoc6,892 bytes../../../.claude-plugin/skills/mempalace-task/SKILL.mddoc7,432 bytes../../../.claude-plugin/skills/mempalace/SKILL.mddoc8,647 bytes../../../.codex-plugin/hooks.jsonasset712 bytes../../../examples/antigravity/hooks.jsonasset436 bytes../../../examples/cursor/hooks.jsonasset426 bytes
Read SKILL.md
# MemPalace
A searchable memory palace for AI — mine projects and conversations, then search them semantically. Verbatim storage, local-first, zero external API by default.
## Prerequisites
Ensure `mempalace` is installed:
```bash
mempalace --version
```
If not installed (uv recommended):
mempalace-recall
When your agent uses it
Recall protocol for MemPalace — search the palace before answering about past work, people, projects, or prior decisions. Apply when the user asks what was decided, what happened before, who someone is, what was discussed last time, or anything that may already be filed in their memory palace; or when mempalace-recall is invoked. Complements the mempalace setup skill and requires the mempalace-mcp server.
What it loads into context
Frontmatter
98 tokensAlways available to the agent
SKILL.md
1,525 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
- Step 0 — Verify MemPalace is available
- Identity
- When to recall
- Protocol
- Tool selection
- Unhappy paths
- Anti-patterns — never do these
- Official References
Files
SKILL.mdcore6,278 bytesSKILL.mdcore6,278 bytesSKILL.mdcore6,278 bytes
Read SKILL.md
# MemPalace Recall
Search-before-answer protocol for MemPalace. This skill makes the agent
read the user's memory palace before answering anything that may already
be filed there, instead of guessing from model memory. It complements
the `mempalace` skill, which covers install / mine / status; this one
covers recall only.
## Step 0 — Verify MemPalace is available
Before relying on recall, confirm MemPalace is installed and reachable:
- Official release page: <https://github.com/MemPalace/mempalace/releases>
- Check installed: `mempalace --version`mempalace
When your agent uses it
Install, configure, and operate MemPalace, including a private local palace, a shared-brain hub, or a client joining an existing hub. Use for first-time setup, MCP wiring, mining, status, palace audit and repair, wings, rooms, drawers, shared-brain identity, or logstream readiness.
What it loads into context
Frontmatter
72 tokensAlways available to the agent
SKILL.md
2,051 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
- Setup protocol
- Palace health: audit and repair session
- Recalling past work
- Cursor-specific notes
- Canonical references
Files
SKILL.mdcore8,647 bytesSKILL.mdcore8,647 bytesSKILL.mdcore8,647 bytesSKILL.mdcore8,647 bytesSKILL.mdcore8,647 bytesSKILL.mdcore8,647 bytesSKILL.mdcore8,647 bytes
Read SKILL.md
# MemPalace Setup
A guided, skill-first setup for a searchable memory palace. The user may have
installed this skill with `npx skills add` before the MemPalace Python package
or MCP server exists; that is the normal bootstrap path.
## Setup protocol
### 1. Inspect before changing anything
- Detect the OS and current agent harness.
- Run `mempalace --version`, `uv --version`, and an appropriate Python version
check. Do not assume that an installed Python package is reachable on PATH.
- Check for an existing palace and MCP registration. Never reinitialize ormempalace-recall
When your agent uses it
Recall protocol for MemPalace — search the palace before answering about past work, people, projects, or prior decisions. Apply when the user asks what was decided, what happened before, who someone is, what was discussed last time, or anything that may already be filed in their memory palace; or when mempalace-recall is invoked. Complements the mempalace setup skill and requires the mempalace-mcp server.
What it loads into context
Frontmatter
98 tokensAlways available to the agent
SKILL.md
1,699 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
- Step 0 — Verify MemPalace is available
- Identity
- When to recall
- Protocol
- Tool selection
- Unhappy paths
- Anti-patterns — never do these
- Official References
Files
SKILL.mdcore6,892 bytesSKILL.mdcore6,892 bytesSKILL.mdcore6,892 bytesSKILL.mdcore6,892 bytes
Read SKILL.md
# MemPalace Recall
Search-before-answer protocol for MemPalace. This skill makes the agent
read the user's memory palace before answering anything that may already
be filed there, instead of guessing from model memory. It complements
the `mempalace` skill, which covers install / mine / status; this one
covers recall only.
## Step 0 — Verify MemPalace is available
Before relying on recall, confirm MemPalace is installed and reachable:
- Official release page: <https://github.com/MemPalace/mempalace/releases>
- Check installed: `mempalace --version`mempalace-task
When your agent uses it
Create, hand off, claim, execute, and close agent tasks through the MemPalace logstream. Use when the user wants to delegate work, prepare a ready-to-paste task for another agent, receive a MemPalace task id, or explicitly launch a supported headless agent in controlled mode.
What it loads into context
Frontmatter
70 tokensAlways available to the agent
SKILL.md
1,713 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
- Verify the coordination seam
- Create a task
- Receive and execute a pasted task
- Controlled headless mode
- Unhappy paths
Files
SKILL.mdcore7,432 bytesSKILL.mdcore7,432 bytesSKILL.mdcore7,432 bytesSKILL.mdcore7,432 bytes
Read SKILL.md
# MemPalace Task
A quick-start workflow for active work moving between agents. Tasks belong in
the logstream, not in memory drawers. The canonical lifecycle and watcher
discipline live in the public
[coordination protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md).
## Verify the coordination seam
1. Confirm `mempalace --version` succeeds.
Also confirm `mempalace task --help` and `mempalace_task_create` exist; if
either is missing, do not assume the older runtime has update-awareness
commands. Explain that it is incompatible, hand off to the `mempalace` setup
skill, and propose the package-manager-appropriate upgrade (`uv tool upgradeaudit
When your agent uses it
Score how well organized the MemPalace palace is, then run a guided repair session — one decision at a time.
What it loads into context
Frontmatter
36 tokensAlways available to the agent
SKILL.md
70 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
Files
SKILL.mdcore309 bytesSKILL.mdcore309 bytes
Read SKILL.md
# MemPalace Audit
Run the following command and follow the returned instructions step by step:
```bash
mempalace instructions audit
```help
When your agent uses it
Show MemPalace help — available commands, usage tips, and getting started guidance.
What it loads into context
Frontmatter
29 tokensAlways available to the agent
SKILL.md
63 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
Files
SKILL.mdcore281 bytesSKILL.mdcore281 bytes
Read SKILL.md
# MemPalace Help
Run the following command and follow the returned instructions step by step:
```bash
mempalace instructions help
```init
When your agent uses it
Initialize a new MemPalace — guided setup for your AI memory palace with ChromaDB backend.
What it loads into context
Frontmatter
36 tokensAlways available to the agent
SKILL.md
70 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
Files
SKILL.mdcore301 bytesSKILL.mdcore301 bytes
Read SKILL.md
# MemPalace Init
Run the following command and follow the returned instructions step by step:
```bash
mempalace instructions init
```mine
When your agent uses it
Mine a project or conversation into your MemPalace — extract and store memories for later retrieval.
What it loads into context
Frontmatter
36 tokensAlways available to the agent
SKILL.md
70 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
Files
SKILL.mdcore310 bytesSKILL.mdcore310 bytes
Read SKILL.md
# MemPalace Mine
Run the following command and follow the returned instructions step by step:
```bash
mempalace instructions mine
```search
When your agent uses it
Search your MemPalace — semantic search across all mined memories, projects, and conversations.
What it loads into context
Frontmatter
30 tokensAlways available to the agent
SKILL.md
64 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
Files
SKILL.mdcore299 bytesSKILL.mdcore299 bytes
Read SKILL.md
# MemPalace Search
Run the following command and follow the returned instructions step by step:
```bash
mempalace instructions search
```status
When your agent uses it
Show MemPalace status — room counts, storage usage, and palace health.
What it loads into context
Frontmatter
28 tokensAlways available to the agent
SKILL.md
62 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
Files
SKILL.mdcore274 bytesSKILL.mdcore274 bytes
Read SKILL.md
# MemPalace Status
Run the following command and follow the returned instructions step by step:
```bash
mempalace instructions status
```mempalace
When your agent uses it
Recall past work or facts; save memories; update relationships; set up or troubleshoot MemPalace in OpenClaw.
What it loads into context
Frontmatter
148 tokensAlways available to the agent
SKILL.md
1,172 tokensWhen the skill is activated
Reference documents
1,809 tokensLoaded on request
Token counts are estimates. Scripts run separately; they are not counted as reference context.
The playbook
- License
Files
SKILL.mdcore5,045 bytesSKILL.mdcore5,045 bytesSKILL.mdcore5,045 bytesSKILL.mdcore5,045 bytesSKILL.mdcore5,045 bytesreferences/setup.mddoc4,976 bytesreferences/tool-selection.mddoc2,997 bytes
Read SKILL.md
# MemPalace
1. **Choose the branch.** For past work, people, preferences, decisions, or
continuity, use the procedure below ([canonical recall][recall]). Honor
session-start context and known wing scope; omit uncertain filters. Skip
recall for greenfield work. Use exposed tools with current schemas/permissions;
logical names below may be namespaced. Reuse an existing connection without
requiring a local CLI or another palace. Never substitute local data on a
remote-only failure. Consult
[tool selection]({baseDir}/references/tool-selection.md) only for unfamiliar
syntax/edge cases; ordinary recall needs no extra read.
For explicitly requested installation or connection diagnosis,
read [setup]({baseDir}/references/setup.md) instead. Proceed once the question
and authorized operation are clear; do not run setup as a recall prerequisite.mempalace
When your agent uses it
Install, configure, and operate MemPalace, including a private local palace, a shared-brain hub, or a client joining an existing hub. Use for first-time setup, MCP wiring, mining, status, palace audit and repair, wings, rooms, drawers, shared-brain identity, or logstream readiness.
What it loads into context
Frontmatter
72 tokensAlways available to the agent
SKILL.md
2,051 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
- Setup protocol
- Palace health: audit and repair session
- Recalling past work
- Cursor-specific notes
- Canonical references
Files
SKILL.mdcore8,647 bytesSKILL.mdcore8,647 bytesSKILL.mdcore8,647 bytesSKILL.mdcore8,647 bytesSKILL.mdcore8,647 bytesSKILL.mdcore8,647 bytesSKILL.mdcore8,647 bytes
Read SKILL.md
# MemPalace Setup
A guided, skill-first setup for a searchable memory palace. The user may have
installed this skill with `npx skills add` before the MemPalace Python package
or MCP server exists; that is the normal bootstrap path.
## Setup protocol
### 1. Inspect before changing anything
- Detect the OS and current agent harness.
- Run `mempalace --version`, `uv --version`, and an appropriate Python version
check. Do not assume that an installed Python package is reachable on PATH.
- Check for an existing palace and MCP registration. Never reinitialize ormempalace-recall
When your agent uses it
Recall protocol for MemPalace — search the palace before answering about past work, people, projects, or prior decisions. Apply when the user asks what was decided, what happened before, who someone is, what was discussed last time, or anything that may already be filed in their memory palace; or when mempalace-recall is invoked. Complements the mempalace setup skill and requires the mempalace-mcp server.
What it loads into context
Frontmatter
98 tokensAlways available to the agent
SKILL.md
1,699 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
- Step 0 — Verify MemPalace is available
- Identity
- When to recall
- Protocol
- Tool selection
- Unhappy paths
- Anti-patterns — never do these
- Official References
Files
SKILL.mdcore6,892 bytesSKILL.mdcore6,892 bytesSKILL.mdcore6,892 bytesSKILL.mdcore6,892 bytes
Read SKILL.md
# MemPalace Recall
Search-before-answer protocol for MemPalace. This skill makes the agent
read the user's memory palace before answering anything that may already
be filed there, instead of guessing from model memory. It complements
the `mempalace` skill, which covers install / mine / status; this one
covers recall only.
## Step 0 — Verify MemPalace is available
Before relying on recall, confirm MemPalace is installed and reachable:
- Official release page: <https://github.com/MemPalace/mempalace/releases>
- Check installed: `mempalace --version`mempalace-task
When your agent uses it
Create, hand off, claim, execute, and close agent tasks through the MemPalace logstream. Use when the user wants to delegate work, prepare a ready-to-paste task for another agent, receive a MemPalace task id, or explicitly launch a supported headless agent in controlled mode.
What it loads into context
Frontmatter
70 tokensAlways available to the agent
SKILL.md
1,713 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
- Verify the coordination seam
- Create a task
- Receive and execute a pasted task
- Controlled headless mode
- Unhappy paths
Files
SKILL.mdcore7,432 bytesSKILL.mdcore7,432 bytesSKILL.mdcore7,432 bytesSKILL.mdcore7,432 bytes
Read SKILL.md
# MemPalace Task
A quick-start workflow for active work moving between agents. Tasks belong in
the logstream, not in memory drawers. The canonical lifecycle and watcher
discipline live in the public
[coordination protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md).
## Verify the coordination seam
1. Confirm `mempalace --version` succeeds.
Also confirm `mempalace task --help` and `mempalace_task_create` exist; if
either is missing, do not assume the older runtime has update-awareness
commands. Explain that it is incompatible, hand off to the `mempalace` setup
skill, and propose the package-manager-appropriate upgrade (`uv tool upgradeSkills included
- mempalace
- mempalace-recall
- mempalace
- mempalace-recall
- mempalace-task
- audit
- help
- init
- mine
- search
- status
- mempalace
- mempalace
- mempalace-recall
- mempalace-task
Repository facts
- License
- MIT
- Created
- Apr 5, 2026
- Last pushed
- Nov 5, 2026
- Forks
- 7,655
- Open issues
- 378
- Project activity
- Active
Rank over the last 12 weeks
Now #32 · best #25 · down 5 in 30 days
- #28 on Oct 9
- #29 from Oct 10 to Oct 21
- #31 from Oct 22 to Oct 24
- #33 on Oct 25
- #31 from Oct 28 to Nov 4
- #32 since Nov 5
5 earlier periods
- #24 on Aug 2
- #24 from Aug 7 to Aug 11
- #25 from Aug 12 to Aug 16
- #26 from Aug 17 to Sep 11
- #27 from Sep 12 to Oct 8
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 50 notes
.antigravity-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.antigravity-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.antigravity-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.antigravity-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.antigravity-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.antigravity-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.antigravity-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.antigravity-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.antigravity-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.antigravity-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.antigravity-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.claude-plugin/hooks/hooks.json:1Analysis incomplete: Referenced executable target unavailable; not analyzed.claude-plugin/hooks/hooks.json:1Analysis incomplete: Referenced executable target unavailable; not analyzed.claude-plugin/hooks/hooks.json:1Analysis incomplete: Referenced executable target unavailable; not analyzed.antigravity-plugin/skills/mempalace-recall/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.antigravity-plugin/skills/mempalace-recall/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.claude-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.claude-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.claude-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed.claude-plugin/skills/mempalace/SKILL.md:1Analysis incomplete: Referenced executable target unavailable; not analyzed
- 12 broad permissionsin 12 files: .claude-plugin/commands/audit.md:1 .claude-plugin/commands/help.md:1 .claude-plugin/commands/init.md:1 and 9 more
- No dangerous shell commands
- 102 network referencesin 13 files: .claude-plugin/skills/mempalace-recall/SKILL.md:18 .claude-plugin/skills/mempalace/SKILL.md:78 mempalace-recall/SKILL.md:18 and 10 more
- No hidden or obfuscated text
- No instruction overrides
- 4 executable scriptsin 4 files: .claude-plugin/hooks/mempal-precompact-hook.sh:1 .claude-plugin/hooks/mempal-session-end-hook.sh:1 .claude-plugin/hooks/mempal-stop-hook.sh:1 and 1 more
- No access to secrets
- No requests to vote in Hot
Show all 118 signals by file
.antigravity-plugin/skills/mempalace-recall/SKILL.md
- Network accessline 18
- Official release page: <https://github.com/MemPalace/mempalace/releases> - Network accessline 58
facts, `mempalace_search` for source text, or `mempalace_diary_read` for recent agent continuity. Stop when answered; follow the canonical [query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-pr - Network accessline 59
recent agent continuity. Stop when answered; follow the canonical [query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-protocol.md#retrieve-only-the-context-you-need). 3. **If unsure** about a fact - Network accessline 60
[query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-protocol.md#retrieve-only-the-context-you-need). - Network accessline 75
The full canonical protocol — shared verbatim with the Antigravity recall rule and the other integrations — lives in [`integrations/shared/recall-protocol.md`](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/recall-prot - Network accessline 77
[`integrations/shared/recall-protocol.md`](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/recall-protocol.md). - Network accessline 123
- MemPalace: <https://github.com/MemPalace/mempalace> - Network accessline 124
- MemPalace releases: <https://github.com/MemPalace/mempalace/releases> - Network accessline 125
- Antigravity documentation: <https://antigravity.google/docs> - Network accessline 126
- Agent Skills specification: <https://agentskills.io/specification>
.claude-plugin/commands/audit.md
- Broad permissionsline 1
Bash, Read
.claude-plugin/commands/help.md
- Broad permissionsline 1
Bash, Read
.claude-plugin/commands/init.md
- Broad permissionsline 1
Bash, Read, Write, Edit, Glob, Grep
.claude-plugin/commands/mine.md
- Broad permissionsline 1
Bash, Read, Write, Edit, Glob, Grep
.claude-plugin/commands/search.md
- Broad permissionsline 1
Bash, Read
.claude-plugin/commands/status.md
- Broad permissionsline 1
Bash, Read
.claude-plugin/hooks/mempal-precompact-hook.sh
- Executable scriptline 1
#!/bin/bash
.claude-plugin/hooks/mempal-session-end-hook.sh
- Executable scriptline 1
#!/bin/bash - Network accessline 4
# Claude Code documents a default SessionEnd hook timeout of 1.5s, and # "timeouts set on plugin-provided hooks do not raise the budget" # (https://code.claude.com/docs/en/hooks). A cold `mempalace` start alone - Network accessline 5
# "timeouts set on plugin-provided hooks do not raise the budget" # (https://code.claude.com/docs/en/hooks). A cold `mempalace` start alone # exceeds 1.5s, so the final mine must NOT run in the foreground: it would be - Network accessline 6
# (https://code.claude.com/docs/en/hooks). A cold `mempalace` start alone
.claude-plugin/hooks/mempal-stop-hook.sh
- Executable scriptline 1
#!/bin/bash
.claude-plugin/hooks/mempal_hook_run.sh
- Executable scriptline 1
#!/bin/bash
.claude-plugin/marketplace.json
.claude-plugin/plugin.json
.claude-plugin/skills/mempalace-recall/SKILL.md
- Network accessline 18
- Official release page: <https://github.com/MemPalace/mempalace/releases> - Network accessline 18
- Official release page: <https://github.com/MemPalace/mempalace/releases> - Network accessline 57
facts, `mempalace_search` for source text, or `mempalace_diary_read` for recent agent continuity. Stop when answered; follow the canonical [query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-pr - Network accessline 57
facts, `mempalace_search` for source text, or `mempalace_diary_read` for recent agent continuity. Stop when answered; follow the canonical [query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-pr - Network accessline 58
recent agent continuity. Stop when answered; follow the canonical [query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-protocol.md#retrieve-only-the-context-you-need). 3. If unsure about a fact: sa - Network accessline 58
recent agent continuity. Stop when answered; follow the canonical [query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-protocol.md#retrieve-only-the-context-you-need). 3. If unsure about a fact: sa - Network accessline 59
[query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-protocol.md#retrieve-only-the-context-you-need). - Network accessline 59
[query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-protocol.md#retrieve-only-the-context-you-need). - Network accessline 70
The full canonical protocol — shared verbatim with the Cursor recall rule and the other integrations — is published in the [MemPalace repository](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/recall-protocol.md). - Network accessline 70
The full canonical protocol — shared verbatim with the Cursor recall rule and the other integrations — is published in the [MemPalace repository](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/recall-protocol.md). - Network accessline 72
[MemPalace repository](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/recall-protocol.md). - Network accessline 72
[MemPalace repository](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/recall-protocol.md). - Network accessline 93
`mempalace_patch_submit`, `mempalace_artifact_get`), not drawers or search. The canonical protocol is published in the [MemPalace repository](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md). - Network accessline 93
`mempalace_patch_submit`, `mempalace_artifact_get`), not drawers or search. The canonical protocol is published in the [MemPalace repository](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md). - Network accessline 95
[MemPalace repository](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md). - Network accessline 95
[MemPalace repository](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md). - Network accessline 138
- MemPalace: <https://github.com/MemPalace/mempalace> - Network accessline 138
- MemPalace: <https://github.com/MemPalace/mempalace> - Network accessline 139
- MemPalace releases: <https://github.com/MemPalace/mempalace/releases> - Network accessline 139
- MemPalace releases: <https://github.com/MemPalace/mempalace/releases> - Network accessline 140
- Cursor Skills documentation: <https://cursor.com/docs/skills> - Network accessline 140
- Cursor Skills documentation: <https://cursor.com/docs/skills> - Network accessline 141
- Agent Skills specification: <https://agentskills.io/specification> - Network accessline 141
- Agent Skills specification: <https://agentskills.io/specification>
.claude-plugin/skills/mempalace-task/SKILL.md
- Network accessline 9
the logstream, not in memory drawers. The canonical lifecycle and watcher discipline live in the public [coordination protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md). - Network accessline 9
the logstream, not in memory drawers. The canonical lifecycle and watcher discipline live in the public [coordination protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md). - Network accessline 11
[coordination protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md). - Network accessline 11
[coordination protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md).
.claude-plugin/skills/mempalace/SKILL.md
- Network accessline 78
For a shared-brain hub, guide the user through `mempalace serve` and the [official shared-brain guide](https://mempalaceofficial.com/guide/shared-brain.html). Do not expose a non-loopback server without - Network accessline 78
For a shared-brain hub, guide the user through `mempalace serve` and the [official shared-brain guide](https://mempalaceofficial.com/guide/shared-brain.html). Do not expose a non-loopback server without - Network accessline 79
[official shared-brain guide](https://mempalaceofficial.com/guide/shared-brain.html). Do not expose a non-loopback server without - Network accessline 79
[official shared-brain guide](https://mempalaceofficial.com/guide/shared-brain.html). Do not expose a non-loopback server without - Network accessline 175
- The Cursor plugin auto-registers `mempalace-mcp`; a standalone `npx skills add` installation does not. Always verify the live tool list. - For automatic background saving every N agent turns plus session-start memory recall, also instal - Network accessline 175
- The Cursor plugin auto-registers `mempalace-mcp`; a standalone `npx skills add` installation does not. Always verify the live tool list. - For automatic background saving every N agent turns plus session-start memory recall, also instal - Network accessline 176
add` installation does not. Always verify the live tool list. - For automatic background saving every N agent turns plus session-start memory recall, also install the Cursor hooks separately by running `hooks/cursor/install.sh --scope user` - Network accessline 176
add` installation does not. Always verify the live tool list. - For automatic background saving every N agent turns plus session-start memory recall, also install the Cursor hooks separately by running `hooks/cursor/install.sh --scope user` - Network accessline 177
- For automatic background saving every N agent turns plus session-start memory recall, also install the Cursor hooks separately by running `hooks/cursor/install.sh --scope user` from a cloned MemPalace repo. See the [Cursor hooks guide](ht - Network accessline 177
- For automatic background saving every N agent turns plus session-start memory recall, also install the Cursor hooks separately by running `hooks/cursor/install.sh --scope user` from a cloned MemPalace repo. See the [Cursor hooks guide](ht - Network accessline 182
- [Shared brain](https://mempalaceofficial.com/guide/shared-brain.html) - Network accessline 182
- [Shared brain](https://mempalaceofficial.com/guide/shared-brain.html) - Network accessline 183
- [Coordination protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md) - Network accessline 183
- [Coordination protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md) - Network accessline 184
- [Recall protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/recall-protocol.md) - Network accessline 184
- [Recall protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/recall-protocol.md)
.codex-plugin/skills/audit/SKILL.md
- Broad permissionsline 1
Bash, Read
.codex-plugin/skills/help/SKILL.md
- Broad permissionsline 1
Bash, Read
.codex-plugin/skills/init/SKILL.md
- Broad permissionsline 1
Bash, Read, Write, Edit
.codex-plugin/skills/mine/SKILL.md
- Broad permissionsline 1
Bash, Read, Glob, Grep
.codex-plugin/skills/search/SKILL.md
- Broad permissionsline 1
Bash, Read
.codex-plugin/skills/status/SKILL.md
- Broad permissionsline 1
Bash, Read
integrations/openclaw/SKILL.md
- Network accessline 3
description: "Recall past work or facts; save memories; update relationships; set up or troubleshoot MemPalace in OpenClaw." version: 3.12.0 homepage: https://github.com/MemPalace/mempalace - Network accessline 4
version: 3.12.0 homepage: https://github.com/MemPalace/mempalace user-invocable: true - Network accessline 5
homepage: https://github.com/MemPalace/mempalace - Network accessline 82
[recall]: https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-protocol.md#retrieve-only-the-context-you-need - Network accessline 83
[recovery]: https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-protocol.md#recovering-a-corrupt-index - Network accessline 87
[MemPalace](https://github.com/MemPalace/mempalace) is MIT licensed. Created by Milla Jovovich, Ben Sigman, Igor Lins e Silva, and contributors.
integrations/openclaw/references/setup.md
- Network accessline 41
Consult running-version OpenClaw docs; public equivalents: [MCP](https://docs.openclaw.ai/tools/mcp), - Network accessline 42
[MCP](https://docs.openclaw.ai/tools/mcp), - Network accessline 43
[MCP registry](https://docs.openclaw.ai/cli/mcp/registry), and - Network accessline 44
[Skills](https://docs.openclaw.ai/tools/skills). Inspect - Network accessline 64
openclaw mcp set mempalace '{"url":"http://127.0.0.1:8765/mcp","transport":"streamable-http"}' - Network accessline 99
[hub]: https://mempalaceofficial.com/guide/shared-brain.html#_1-start-the-hub - Network accessline 100
[remote]: https://mempalaceofficial.com/guide/remote-server.html#_3-serve-mcp-over-http - Network accessline 101
[install]: https://github.com/MemPalace/mempalace/blob/develop/skills/mempalace/SKILL.md#2-install-the-cli-when-necessary - Network accessline 102
[routing]: https://github.com/MemPalace/mempalace/blob/develop/docs/cli-write-routing.md#safety-properties
integrations/openclaw/references/tool-selection.md
skills/mempalace-recall/SKILL.md
- Network accessline 18
- Official release page: <https://github.com/MemPalace/mempalace/releases> - Network accessline 57
facts, `mempalace_search` for source text, or `mempalace_diary_read` for recent agent continuity. Stop when answered; follow the canonical [query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-pr - Network accessline 58
recent agent continuity. Stop when answered; follow the canonical [query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-protocol.md#retrieve-only-the-context-you-need). 3. If unsure about a fact: sa - Network accessline 59
[query guide](https://github.com/MemPalace/mempalace/blob/develop/integrations/shared/recall-protocol.md#retrieve-only-the-context-you-need). - Network accessline 70
The full canonical protocol — shared verbatim with the Cursor recall rule and the other integrations — is published in the [MemPalace repository](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/recall-protocol.md). - Network accessline 72
[MemPalace repository](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/recall-protocol.md). - Network accessline 93
`mempalace_patch_submit`, `mempalace_artifact_get`), not drawers or search. The canonical protocol is published in the [MemPalace repository](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md). - Network accessline 95
[MemPalace repository](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md). - Network accessline 138
- MemPalace: <https://github.com/MemPalace/mempalace> - Network accessline 139
- MemPalace releases: <https://github.com/MemPalace/mempalace/releases> - Network accessline 140
- Cursor Skills documentation: <https://cursor.com/docs/skills> - Network accessline 141
- Agent Skills specification: <https://agentskills.io/specification>
skills/mempalace-task/SKILL.md
- Network accessline 9
the logstream, not in memory drawers. The canonical lifecycle and watcher discipline live in the public [coordination protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md). - Network accessline 11
[coordination protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md).
skills/mempalace/SKILL.md
- Network accessline 78
For a shared-brain hub, guide the user through `mempalace serve` and the [official shared-brain guide](https://mempalaceofficial.com/guide/shared-brain.html). Do not expose a non-loopback server without - Network accessline 79
[official shared-brain guide](https://mempalaceofficial.com/guide/shared-brain.html). Do not expose a non-loopback server without - Network accessline 175
- The Cursor plugin auto-registers `mempalace-mcp`; a standalone `npx skills add` installation does not. Always verify the live tool list. - For automatic background saving every N agent turns plus session-start memory recall, also instal - Network accessline 176
add` installation does not. Always verify the live tool list. - For automatic background saving every N agent turns plus session-start memory recall, also install the Cursor hooks separately by running `hooks/cursor/install.sh --scope user` - Network accessline 177
- For automatic background saving every N agent turns plus session-start memory recall, also install the Cursor hooks separately by running `hooks/cursor/install.sh --scope user` from a cloned MemPalace repo. See the [Cursor hooks guide](ht - Network accessline 182
- [Shared brain](https://mempalaceofficial.com/guide/shared-brain.html) - Network accessline 183
- [Coordination protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/coordination-protocol.md) - Network accessline 184
- [Recall protocol](https://github.com/MemPalace/mempalace/blob/main/integrations/shared/recall-protocol.md)
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: 112 facts
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:18Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:58Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:59Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:60Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:75Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:77Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:123Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:124Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:125Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:126Low confidence
- Broad permissions in prose of the plugin configuration.claude-plugin/commands/audit.md:1High confidence
- Broad permissions in prose of the plugin configuration.claude-plugin/commands/help.md:1High confidence
- Broad permissions in prose of the plugin configuration.claude-plugin/commands/init.md:1High confidence
- Broad permissions in prose of the plugin configuration.claude-plugin/commands/mine.md:1High confidence
- Broad permissions in prose of the plugin configuration.claude-plugin/commands/search.md:1High confidence
- Broad permissions in prose of the plugin configuration.claude-plugin/commands/status.md:1High confidence
- Executable script in code of a file that runs.claude-plugin/hooks/mempal-precompact-hook.sh:1High confidence
- Executable script in code of a file that runs.claude-plugin/hooks/mempal-session-end-hook.sh:1High confidence
- Network access in a comment of a file that runs.claude-plugin/hooks/mempal-session-end-hook.sh:4Low confidence
- Network access in a comment of a file that runs.claude-plugin/hooks/mempal-session-end-hook.sh:5Low confidence
- Network access in a comment of a file that runs.claude-plugin/hooks/mempal-session-end-hook.sh:6Low confidence
- Executable script in code of a file that runs.claude-plugin/hooks/mempal-stop-hook.sh:1High confidence
- Executable script in code of a file that runs.claude-plugin/hooks/mempal_hook_run.sh:1High confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-recall/SKILL.md:18Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:18Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-recall/SKILL.md:57Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:57Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-recall/SKILL.md:58Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:58Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-recall/SKILL.md:59Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:59Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-recall/SKILL.md:70Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:70Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-recall/SKILL.md:72Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:72Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-recall/SKILL.md:93Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:93Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-recall/SKILL.md:95Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:95Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-recall/SKILL.md:138Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:138Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-recall/SKILL.md:139Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:139Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-recall/SKILL.md:140Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:140Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-recall/SKILL.md:141Low confidence
- Network access in prose of the SKILL.mdmempalace-recall/SKILL.md:141Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-task/SKILL.md:9Low confidence
- Network access in prose of the SKILL.mdmempalace-task/SKILL.md:9Low confidence
- Network access in prose of the SKILL.md.claude-plugin/skills/mempalace-task/SKILL.md:11Low confidence
and 62 more
Works with
| Agent | Status | Evidence | Install to |
|---|---|---|---|
| Claude Code | Native | Manifest .claude-plugin/marketplace.json | .claude/skills/<skill-name>/ |
| Codex | Native | Manifest AGENTS.md | .agents/skills/<skill-name>/ |
| Cursor | Compatible | Standard SKILL.md .antigravity-plugin/skills/mempalace/SKILL.md | .cursor/skills/<skill-name>/ |
| Gemini CLI | Compatible | Standard SKILL.md .antigravity-plugin/skills/mempalace/SKILL.md | .gemini/skills/<skill-name>/ |
| GitHub Copilot | Compatible | Standard SKILL.md .antigravity-plugin/skills/mempalace/SKILL.md | .github/skills/<skill-name>/ |
| OpenCode | Compatible | Standard SKILL.md .antigravity-plugin/skills/mempalace/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/mempalace-mempalace/)