planning-with-files@3.29.0
CAUTION· install onlyPublisher claims: "Persistent file-based planning for multi-step AI-agent work. Keeps task_plan.md, findings.md, and progress.md on disk; lifecycle hooks inject selected project planning context. Automatic recovery reads project planning files only. Explicit session-catchup.py --metadata reads same-project local agent" Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Bundled scripts ran with no arguments; no agent task scenario for this bundle.- Tested
- 8 of 14 bundled scripts run with no arguments (the rest were not run: per-bundle cap), with decoy credentials in the environment · every file read, file written and host contacted by what ran
- Not tested
- use of the skill by an agent
What changed since 3.28.2
vet-skills diff planning-with-files 3.28.2 3.29.0In the sandbox, this version reached for nothing 3.28.2 did not. Compared from the two recorded runs, not from source.
The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; bundled scripts ran with no arguments; no agent task scenario for this bundle) on 2026-11-21 and scored it 79/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Fake instruction boundary tag; Authorization escalation language. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 1 s · 2 scenariosStatic findings
4. Issue Claude Code's native `/loop <interval> <prompt>` (CC primitive, always available).Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.# above already does it; a non-zero status would kill callers running underFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.# chmod intentionally changes st_mode. Freeze the directory object acrossFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.# that operation, then use the post-chmod identity for every later check.Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.The mode is set by writing a `.mode` file next to the plan (`.planning/<id>/.mode`, or `./.mode` in legacy root mode). `init-session` writes it for you when you pass `--autonomous` or `--gated`.Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.# permanently kill injection at this root over a plan that cannotFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.# a copied project tree) would otherwise kill injection permanently with noFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.chmod =rw "${temp_file}" 2>/dev/null || trueFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.result = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.# set -u is active: an unset IFS or PATH must not kill the hook.Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.chmod 700 "$SNAP_ROOT" 2>/dev/null || :Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.| `npx skills add OthmanAdi/planning-with-files` (or ClawHub) | SKILL.md, scripts, templates only | No, follow the manual fallback below |Fix: Pin versions and sources; vendor anything fetched at install time.Cost to the agent
Tokens estimated at four characters each. Skills load their instruction text when triggered; the figure is what that costs.
Known vulnerabilities
Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.
What would make this Verified
1 point to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Unclaimed publisher≈ +8 on the totalClaim othmanadi: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Touches files outside the project≈ +7 on the totalKeep reads and writes inside the working directory, or declare the exact paths in `filesystem:` and explain why in the README.
- Process kill command≈ +2 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- Python subprocess execution detected≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
- File permission modification command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.
- npx invocation of non-well-known package≈ +2 on the totalPin versions and sources; vendor anything fetched at install time.