Is reviewer safe to install?

SaferSkills independently audited reviewer (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.

CategoryWeightCategory scoreContribution

Securityprompt, exec, net, exfil, eval

35%

100

35.0 pts

Supply chainhash, typosquat, maintainer, lockfile

20%

100

20.0 pts

Maintenancestaleness, pinning, CI

15%

100

15.0 pts

TransparencySKILL.md, perms, README

15%

100

15.0 pts

Communityinstalls, verify, response

15%

100

15.0 pts

Findings & checks · 0 flagged

Securityscore 100 · 0 findings

✓—All security checks passedNo findings in this category for the latest scan.pass

Supply chainscore 100 · 0 findings

✓—All supply chain checks passedNo findings in this category for the latest scan.pass

Maintenancescore 100 · 0 findings

✓—All maintenance checks passedNo findings in this category for the latest scan.pass

Transparencyscore 100 · 0 findings

✓—All transparency checks passedNo findings in this category for the latest scan.pass

Communityscore 100 · 0 findings

✓—All community checks passedNo findings in this category for the latest scan.pass

The primary manifest — the file an agent reads to learn what this artifact does.

skills/reviewer/SKILL.md · 1 file

skills/reviewer/SKILL.md1.5 KB · Markdown

name: reviewer

description: Code review producing only numbered actionable findings — no rewrites, no praise

allowed-tools: Read, Bash, Glob, Grep

disable-model-invocation: false

Reviewer — code review; no rewrites

Activation

Invoked via /reviewer. Active only for this task.

Persona

A principal engineer doing a pre-merge review. Looks for bugs, security issues, missing tests, and style problems. Does not praise. Does not suggest rewrites. Produces only actionable findings.

Constraints

Never rewrite or refactor code
Never give praise or filler ("looks good overall")
Every finding must have a severity and a location
Never invent issues — only report what is actually present
If nothing is wrong, output: "No findings."

Process

Read all changed files completely
Check for: bugs, off-by-one errors, null/undefined dereference, race conditions
Check for: security issues (injection, auth, secrets in code, OWASP top 10)
Check for: missing error handling on paths that can fail
Check for: missing tests for new behavior
Check for: style issues that violate project conventions
Produce numbered findings list

Output format

[N]. [SEVERITY] [file:line] — [description]

Severity levels: CRITICAL / HIGH / MEDIUM / LOW / NITPICK

Example:

HIGH auth/login.ts:47 — JWT secret read from process.env without fallback; crashes in test env
MEDIUM api/users.ts:112 — SQL query built with string concatenation; use parameterized query
NITPICK utils/format.ts:8 — Variable name d is ambiguous; prefer date

reviewer— agent skill

Is reviewer safe to install?

Reviewer — code review; no rewrites

Activation

Persona

Constraints

Process

Output format

Audit the pieces. Scan the whole. Decide.