@openbmb/memory-maintenance
ADiagnose and maintain ClawXMemory index health using scripts and tools. Use when installation, indexing, or data quality looks wrong.
Install
agr install @openbmb/memory-maintenance --target claudeWrites 1 file into .claude/skills/, pinned to git-4a17a351.
- .claude/skills/memory-maintenance/SKILL.md
Document
name: memory-maintenance description: Diagnose and maintain ClawXMemory index health using scripts and tools. Use when installation, indexing, or data quality looks wrong. homepage: https://github.com/OpenBMB/ClawXMemory/tree/main/clawxmemory metadata: {"openclaw":{"skillKey":"openbmb-clawxmemory","requires":{"config":["plugins.entries.openbmb-clawxmemory.enabled"]}}}
Memory Maintenance
Use this skill for diagnosis and maintenance of ClawXMemory storage/indexes.
Mixed Mode Policy
- Prefer tool calls for normal user-facing retrieval flows.
- Use scripts only for diagnostics, data checks, and troubleshooting.
Recommended Workflow
- Read current memory behavior with tools:
memory_overviewmemory_listmemory_searchmemory_get
- If the user wants an immediate refresh or you see pending memory, run
memory_flush. - If behavior is still suspicious, run scripts:
node {baseDir}/scripts/inspect-indexes.mjs --db <path>node {baseDir}/scripts/recent-sessions.mjs --db <path> --limit 5
- Compare script output with tool output.
- Report concrete findings:
- missing data
- stale indexing
- low extraction quality
- query mismatch
Script Defaults
- Default DB path:
~/.openclaw/clawxmemory/memory.sqlite - You can override with
--db /absolute/path/to/memory.sqlite
Common Debug Cases
- Install succeeds but recall empty: start with
memory_overview, then check table counts andlastIndexedAt. - Project answers weak: start with
memory_search, then inspectl2_projectand recentl1ids usingmemory_get. - Timeline mismatch: inspect
l2_timeids frommemory_search, then compare against the latestl0sessions. - Recent turn not visible yet: inspect
pendingL0inmemory_overview, then runmemory_flush.
Guardrails
- Scripts are read-only diagnostics in this skill.
- Do not modify DB directly from this skill unless user explicitly asks for data repair.
Trustgrade A
- passBody integrity
Whether the stored document is plausibly the kind of file the artifact declares, rather than something fetched by mistake.
- passType matchnot applicable to this artifact type
Whether the artifact is really the kind of thing its metadata claims it is.
- passFreshness
How long since the source repository was last pushed to.
- passPrompt injection
Scans the artifact's own text for instructions aimed at your agent rather than at you.
- passLicense
Whether the source repository declares an SPDX license permissive enough to redistribute.
How the grade is calculated
Each check contributes 0 points when it passes, 1 when it warns, and 2 when it fails. The total maps to a letter:
- Aevery check passed
- Bone warning
- Ctwo warnings
- Dprompt injection or body integrity failed, or three warnings
- Fone of those failed, and something else is wrong
These are automated hygiene checks, not a security audit, and not a dependency or vulnerability scan. A grade of A means nothing was flagged — not that the artifact is safe.
Versions
git-4a17a35124e72026-07-31