agentleFS
thoitiettxl-cyberthoitiettxl-cyber/Claw-II/skills/memory-setup/SKILL.md

memory-setup

Skill2 starsChanged 7 months agoMITView on GitHub

Standardize memory search settings for OpenClaw-Module Android according to current schema; Focus on properly configuring fields, memory file layout, and secure verification processes.

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

15 rules. 0 are common, 13 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 13

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Verify memory search operates before concluding done.

    line 30
  2. Distinctive:

    Do not arbitrarily edit the config if the user has not approved the patch.

    line 31
  3. Distinctive:

    Do not use legacy config path/tooling.

    line 35
  4. Distinctive:

    No field instructions do not exist in the current schema.

    line 36
  5. Distinctive:

    Do not hardcode API keys/tokens in the docs.

    line 37
  6. Distinctive:

    Do not run the default network setup if the user has not requested it.

    line 38
  7. Distinctive:

    When editing config: always backup + validate + restart with notes.

    line 39
  8. Very specific:

    Use gateway.config.schema to confirm the memory field is valid.

    line 45
  9. Very specific:

    Use gateway.config.get to see the current status.

    line 46
  10. Distinctive:

    Wait for the restart to complete.

    line 65
  11. Distinctive:

    Verify with an actual memory query.

    line 66
  12. Distinctive:

    If memory is wrong/missing: update the corresponding memory file.

    line 82
  13. Distinctive:

    If input/permissions are missing: report what is missing and clearly request additional information.

    line 127
  14. Distinctive:

    If runtime/tool ​​error: brief error return + possible cause + suggested fix.

    line 128
  15. Distinctive:

    If there is a risk policy/safety: stop and escalate according to approval checkpoint.

    line 129

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

memory-setup (Android Module Refactor)

This skill is refactored to set up memory according to current OpenClaw standards, avoiding drift from the old config structure.

Indexed Reference Map

  • @[SKILL.md#L1-L129] Standard Frontmatter.
  • @[SKILL.md#L12-L129] Scope + safety principles.
  • @[SKILL.md#L28-L129] SOP memory setup according to current schema.
  • @[SKILL.md#L71-L129] Valid Config snippets for OpenClaw.
  • @[references/openclaw-memory-schema-map.md#L1-L39] Mapping field is correct according to the schema.
  • @[references/android-module-memory-layout.md#L1-L28] Consistent layout file Android module.
  • @[scripts/self_audit.sh#L1-L87] Self-audit static cho skill.

Scope

  1. Set agents.defaults.memorySearch to the correct schema.
  2. Normalize MEMORY.md + memory/YYYY-MM-DD.md usage.
  3. Verify memory search operates before concluding done.
  4. Do not arbitrarily edit the config if the user has not approved the patch.

Hard Rules

  • Do not use legacy config path/tooling.
  • No field instructions do not exist in the current schema.
  • Do not hardcode API keys/tokens in the docs.
  • Do not run the default network setup if the user has not requested it.
  • When editing config: always backup + validate + restart with notes.

SOP (Tool-first)

Step A — Inspect current config schema

  • Use gateway.config.schema to confirm the memory field is valid.
  • Use gateway.config.get to see the current status.

Step B — Plan minimal patch

Patch in the correct location:

  • agents.defaults.memorySearch

Commonly used fields:

  • enabled
  • provider (gemini|voyage|openai|local|mistral)
  • model
  • sources (memory, sessions)
  • sync.onSessionStart, sync.watch
  • query.minScore, query.maxResults

Step C — Apply safely (only when approved)

  1. Backup config (or rely on config guardian process).
  2. gateway.config.patch with minimal patch.
  3. Wait for the restart to complete.
  4. Verify with an actual memory query.

Step D — Workspace memory structure

Obligatory:

  • MEMORY.md (curated long-term)
  • memory/YYYY-MM-DD.md (daily notes)

Recommendation:

  • memory/projects/
  • memory/people/
  • memory/system/

Step E — Verification checklist

  • memory_search returns results from MEMORY.md or memory/*.md.
  • If memory is wrong/missing: update the corresponding memory file.
  • Report results + confidence.

Config Snippet (Valid Shape Example)

{
  "agents": {
    "defaults": {
      "memorySearch": {
        "enabled": true,
        "provider": "gemini",
        "model": "models/gemini-embedding-001",
        "sources": ["memory", "sessions"],
        "sync": {
          "onSessionStart": true,
          "watch": true
        },
        "query": {
          "minScore": 0.3,
          "maxResults": 20
        }
      }
    }
  }
}

Output Contract

MEMORY SETUP REVIEW
Config state: <enabled/provider/model>
Schema alignment: <pass/fail>
Memory files: <present/missing>
Verification: <pass/fail>
Notes: <risks + next action>

Trigger Rules

  • Positive: Triggers when the user request matches memory-setup according to description: Standardizes memory search settings for OpenClaw-Module Android according to the current schema; Focus on properly configuring fields, memory file layout, and secure verification processes.
  • Negative: Not enabled for requests outside scope or without the workflow/expertise of this skill.
  • Escalation: If there is an external action/network write or sensitive operation, an approval checkpoint must be added before executing.

Error Handling

  • If input/permissions are missing: report what is missing and clearly request additional information.
  • If runtime/tool ​​error: brief error return + possible cause + suggested fix.
  • If there is a risk policy/safety: stop and escalate according to approval checkpoint.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.