skillit-refine — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited skillit-refine (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
A bulleted imperative like {match} tells the agent to never reveal, disclose, or mention something to the user. Used adversarially it can instruct the agent to hide its tool calls or lie about what it did — stripping the transparency a user relies on to trust the agent.
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.
Autonomously improve an MCP skill via the skillit audit→draft→review loop (build or runtime mode)
Autonomously improve a skill via the audit→draft→review loop
Usage:
[options]| Flag | Type | Required | Default | Env | Description |
|---|---|---|---|---|---|
--mcp | string | yes | — | — | path to mcp.json or MCP config file |
--server | string | no | — | — | server name within the config (defaults to first enabled) |
--overlay | string | no | — | — | path to overlay JSON file (runtime mode only) |
--mode | string | no | — | — | refine mode: build or runtime (auto-detected if omitted) |
--source-glob | string | no | — | — | glob pattern for TypeScript source files (build mode) |
--max-iterations | string | no | 5 | — | iteration cap (default 5) |
--items | string | no | 5 | — | work items per iteration (default 5) |
Load these on demand — do NOT read all at once:
references/commands.md for flags, arguments, and defaults~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.