ClawHub

Openclaw Token Memory Optimizer

v1.0.0

Optimization suite for OpenClaw agents to prevent token leaks and context bloat. Use when an agent needs to implement background task isolation (Cron) or a Reset & Summarize workflow (RAG).

⭐ 0· 1.1k·4 current·4 all-time
by@klautimus
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name and description (token/context optimization, RAG, cron isolation) match the SKILL.md instructions and the included script. All file accesses (openclaw.json, memory files, ~/.openclaw/sessions) are relevant to the documented purpose. No unrelated credentials, binaries, or services are requested.
✓
Instruction Scope
Runtime instructions focus on updating openclaw.json, using local embeddings/memory_search, indexing local session transcripts, and restarting the gateway. The SKILL.md directs reading local config and memory files only and does not instruct transmitting data to external endpoints or accessing unrelated system secrets. It does ask the agent/user to read files under home (expected for this skill).
✓
Install Mechanism
No install spec is provided (instruction-only skill) — lowest-risk distribution. A helper script (scripts/check-optimization.sh) is included but there is no automatic download or execution mechanism. The script is plain shell and relies on common POSIX utilities; nothing is fetched from remote URLs or extracted.
✓
Credentials
The skill declares no environment variables or credentials and indeed does not require any. The script reads local files (config and workspace memory files) which is proportionate to a tool designed to check/advise on local OpenClaw configuration. No unrelated secrets or credentials are referenced.
✓
Persistence & Privilege
always is false and the skill is user-invocable; model invocation is allowed (default) which is normal. The skill does not request persistent system-wide changes or modify other skills' configs. It recommends the user run 'openclaw gateway restart' (a normal operational action).
Assessment
This skill appears coherent and low-risk for its stated purpose, but take these practical precautions before installing or running any included scripts: 1) Review the included scripts yourself (scripts/check-optimization.sh) — they read local config and workspace files but do not exfiltrate data. 2) Back up your openclaw.json and MEMORY.md before making changes. 3) The helper script assumes workspace paths relative to your config (may not match your layout); test it in a safe environment. 4) There is no external network activity in the skill, but if you plan to run commands like 'openclaw gateway restart', ensure you understand their operational impact. 5) If you prefer, copy the SKILL.md guidance into your operational playbook and make changes manually rather than running scripts from untrusted sources.

Like a lobster shell, security has layers — review code before you run it.

latestvk9761hh8yz2w568eqwfy1x4g198101t4

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments