todo-check — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited todo-check (Agent Skill) and scored it 87/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 3 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 3 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
This skill requires Python 3.8+ and standard library only. No external packages needed.
To install this skill's dependencies:
pip-compile ./requirements.in
pip install -r ./requirements.txtSee ./requirements.txt for the dependency lockfile (currently empty — standard library only).
Check for TODOs and debt markers in a file.
python ${CLAUDE_PLUGIN_ROOT}/skills/todo-check/scripts/check_todos.py <path>
<example> Context: Directly auditing a specific file. user: "Are there any todos left in project_logic.py?" assistant: "I'll run the todo-check audit to find any pending work items in project_logic.py." </example>
<example> Context: Scanning for debt as part of a review. user: "List the pending work items in utils.py before we merge." assistant: "I'll use the todo-check tool to scan utils.py for any TODO markers or debt." </example>
<example> Context: Agent proactively audits before proposing a code change. assistant: [autonomously, before suggesting a refactor] "Let me run todo-check on this file first to surface any existing debt markers before adding more changes." <commentary> Implicit audit trigger -- agent uses todo-check as a pre-flight check, no user prompt needed. </commentary> </example>
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.