Is audit safe to install?

SaferSkills independently audited audit (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/audit/SKILL.md · 1 file

skills/audit/SKILL.md14.5 KB · Markdown

name: audit

description: Deep security and performance scan with structured reporting.

version: 3.2.0

model: claude-opus-4-6

/audit Workflow

Inputs

Scope: $ARGUMENTS (optional: "plan", "code", "full")
plan: Audit a plan file before implementation
code: Audit recent uncommitted changes (default)
full: Full codebase scan

Role

You are the audit coordinator. You dispatch security, performance, and QA scans, then synthesize results into actionable reports. You do NOT fix issues yourself — you identify and report them with severity ratings.

Step 1 — Determine scope

Tool: Bash (direct — coordinator does this)

Run: git status --porcelain

Scope resolution:

If $ARGUMENTS is empty:
If git status shows uncommitted changes: scope = "code"
Else: scope = "full"
Else: scope = $ARGUMENTS

Validate scope is one of: plan, code, full. If not, stop with: "Invalid scope. Use: /audit [plan|code|full]"

Derive timestamp: [timestamp] = current ISO datetime (e.g., 2026-02-07T12-30-00)

Initialize audit logging:

Tool: Bash

# --- Audit Logging Setup ---
RUN_ID=$(date +%Y%m%d-%H%M%S)-$(cat /dev/urandom | LC_ALL=C tr -dc 'a-z0-9' | head -c 6)
AUDIT_LOG_DIR="./plans/audit-logs"
mkdir -p "$AUDIT_LOG_DIR"
AUDIT_LOG="$AUDIT_LOG_DIR/audit-${RUN_ID}.jsonl"
STATE_FILE=".audit-audit-state-${RUN_ID}.json"

python3 -c "
import json
state = {
    'run_id': '${RUN_ID}',
    'audit_log': '${AUDIT_LOG}',
    'skill': 'audit',
    'skill_version': '3.2.0',
    'security_maturity': 'advisory',
    'hmac_key': ''
}
with open('${STATE_FILE}', 'w') as f:
    json.dump(state, f)
print('Audit skill state file created: ${STATE_FILE}')
"

bash scripts/emit-audit-event.sh "$STATE_FILE" \
  "{\"event_type\":\"run_start\",\"scope\":\"${AUDIT_SCOPE:-unknown}\"}"

bash scripts/emit-audit-event.sh "$STATE_FILE" \
  '{"event_type":"step_start","step":"step_1_determine_scope","step_name":"Determine scope","agent_type":"coordinator"}'

echo "Audit skill log: $AUDIT_LOG"

Emit step_end for Step 1:

Tool: Bash

bash scripts/emit-audit-event.sh ".audit-audit-state-${RUN_ID}.json" \
  '{"event_type":"step_end","step":"step_1_determine_scope","step_name":"Determine scope","agent_type":"coordinator"}'

Step 2 — Security scan

Emit step_start for Step 2:

Tool: Bash

bash scripts/emit-audit-event.sh ".audit-audit-state-${RUN_ID}.json" \
  '{"event_type":"step_start","step":"step_2_security_scan","step_name":"Security scan","agent_type":"coordinator"}'

Secure-review composability check:

Tool: Glob

Glob for ~/.claude/skills/secure-review/SKILL.md

If found AND scope is NOT "plan":

Output: "Using /secure-review for deep security analysis (composability mode)."
Dispatch /secure-review instead of the built-in security scan.

Tool: Task, subagent_type=general-purpose, model=claude-opus-4-6

Prompt: "You are running a deep security review as part of the /audit workflow.

Read the secure-review skill definition at ~/.claude/skills/secure-review/SKILL.md. Execute its full scanning workflow (vulnerability, data flow, auth/authz scans).

Scope: [map audit scope to secure-review scope: 'code' -> 'changes', 'full' -> 'full']

Write your findings to ./plans/audit-[timestamp].security.md (use the audit naming convention, not the secure-review convention, so the synthesis step can find it).

Include the standard secure-review output: verdict, severity-rated findings, redacted secrets.

CRITICAL: Never include actual secret values. Redact to first 4 / last 4 characters."

Skip the existing built-in security scan below. Proceed to Step 3 (Performance scan).

If not found OR scope is "plan":

If not found: Output: "secure-review skill not deployed. Using built-in security scan."
If scope is "plan": Output: "Scope is 'plan' — using built-in plan security analysis (secure-review scans code, not plans)."
Continue with the existing built-in security scan (unchanged behavior below).

Pre-check: Glob for .claude/agents/security-analyst*.md

Tool: Glob (direct — coordinator does this)

Pattern: .claude/agents/security-analyst*.md

If found: "Using project-specific security-analyst for security scan" If not found: "No project-specific security-analyst found. Using generic Task subagent for security scan. For project-tailored scanning, generate one: gen-agent . --type security-analyst"

Tool: Task, subagent_type=general-purpose, model=claude-opus-4-6

If scope is "plan":

If security-analyst agent found: Prompt: "Read .claude/agents/security-analyst*.md for your role context and scanning frameworks (STRIDE, OWASP Top 10, DREAD, compliance checklists). Then read the plan file at $ARGUMENTS (after 'plan' keyword). Analyze for security risks:
Authentication/authorization gaps
Data exposure risks
Input validation requirements
Cryptographic requirements
Secrets management