repo-visuals-retro — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited repo-visuals-retro (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
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.
The repo-visuals skill scores every run on five criteria (see repo-visuals/SKILL.md §6). This meta-skill turns that accumulated evidence into skill improvements.
./evaluations/runs/ in the user's working directory (before that, the sample is noise).repo-visuals is planned and you want the evidence base first.Do not run this every session. Retros are more valuable with accumulated samples.
repo-visuals skill's SKILL.md — the current skill definition (sibling skill in the same plugin)./evaluations/index.md in the user's working directory — curated aggregate from previous retros./evaluations/runs/*.md in the user's working directory — all per-run raw evaluations since the last retroRead all inputs. Build a small table:
Name each pattern concretely:
For each pattern, consult the relevant expert skill(s) if available in the environment. Examples:
frontend-design skill. Feed it the final HTML + screenshots from low-scoring runs; ask for specific design-rule violations.skill-creator skill. Ask whether the skill's Phase N language is ambiguous or missing a step.evaluate-plugins style skill. Ask whether the 5-criterion scorecard is well-calibrated.Detect which skills are actually available — don't reference non-existent ones.
Write a diff-style proposal: for each pattern, which section of the repo-visuals skill's SKILL.md changes and why. Example:
Pattern: Hero moment delivery weak on CLI tools (2.6/5 avg over 4 runs).
Hypothesis: §1.4a probes don't surface CLI-specific hero moments
(install speed, clear command output, shell ergonomics).
Proposed edit to §1.4a: add a CLI-tool branch to the probing list with
3 tool-specific prompts.Show the proposed diff. Wait for user approval before applying.
On user approval:
repo-visuals skill's SKILL.md with the approved changes./evaluations/index.md (date, patterns identified, edits applied)./evaluations/runs/processed/ so the next retro only sees new samplesrepo-visuals skill SKILL.md./evaluations/index.md with retro summaryrepo-visuals itself~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.