Openclaw Config Guard
v1.0.0Audit and safely repair OpenClaw configuration with deterministic validation, backups, rollback, and change reporting. Use when asked to review or modify `op...
⭐ 1· 180·2 current·2 all-time
by@soal2
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description match behavior: the SKILL.md and the helper script exclusively call the local 'openclaw' CLI, read/write the local config file, make backups, validate, run 'doctor', and produce reports. No unrelated credentials, binaries, or remote services are requested.
Instruction Scope
Runtime instructions are narrowly scoped to locating the config, consulting the documented official sources, running the included Python helper and the OpenClaw CLI, making backups, validating, diffing, and producing a report. The script reads local files and runs local CLI commands only; instructions explicitly prohibit aggressive actions (no automatic doctor --fix, no restart, no wholesale rewrites) and require user approval for non-blocking changes.
Install Mechanism
No install spec; this is instruction-only plus an included helper script. Nothing is downloaded or installed from external URLs, and no archives are extracted.
Credentials
The skill declares no required env vars or credentials. The script temporarily sets OPENCLAW_CONFIG_PATH when invoking the 'openclaw' CLI to point it at a chosen config file; that is proportional and limited to the stated purpose.
Persistence & Privilege
always is false and the skill is user-invocable. The skill does not request permanent elevated presence or modify other skills or system-wide agent settings. Its file writes are limited to creating backup copies of the config.
Assessment
This skill appears to do what it says: run the OpenClaw CLI, back up the config, validate, optionally apply narrowly-justified fixes, and produce a report. Before installing or running it: 1) ensure you have the openclaw CLI locally (the script depends on it); 2) review the helper script yourself if you want extra assurance (it runs subprocesses and reads/writes config files); 3) run it in dry-run/audit mode first (audit/validate/doctor) to see what it would change; 4) keep an eye on the backup path it creates before approving any write; and 5) verify the referenced docs (the SKILL.md lists official docs) if the script recommends a fix. If you need the agent to apply changes, require explicit user confirmation before any write — the skill's instructions already recommend that, but enforcing confirmation protects you from unintended edits.Like a lobster shell, security has layers — review code before you run it.
latestvk971p77qwxdxgwxmdyx5r9dmmn82snqb
License
MIT-0
Free to use, modify, and redistribute. No attribution required.