AC sentinel-vanguard
AI Agent skill security auditor. Use this skill whenever the user wants to audit, review, vet, or assess the safety and security of any AI skill, Claude skill, ClawHub skill, or AI agent tool. Triggers on phrases like "check this skill for safety", "audit this AI tool", "is this skill safe to use", "scan for prompt injection", "review skill security", "check for malicious packages", "vetting a skill", or any request to evaluate trustworthiness of agent-facing code. Also triggers when users paste skill content, code snippets, SKILL.md files, or ask about supply chain risks in AI tooling.
As a process C 55/100 · Has gaps — weak spots: result and completion, inputs and preconditions, running it twice
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 3
✓ No critical or high findings
Medium and low: 3
-
low Dangerous commands
cmd-privilegereferences/l1-rules.md:60Privilege escalation / world-writable permissions (documentation table row; documentation of a security skill)| L1-P07 | `chmod 777` | MEDIUM | World-writable permissions |
tablesecurity skill -
low Dangerous commands
cmd-privilegereferences/l1-rules.md:61Privilege escalation / world-writable permissions (documentation table row; documentation of a security skill)| L1-P08 | `setuid` / `setgid` | CRITICAL | Privilege escalation |
tablesecurity skill -
low Risky intent
intent-offensive-securityreferences/l1-rules.md:61Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)| L1-P08 | `setuid` / `setgid` | CRITICAL | Privilege escalation |
detector
Files scanned: 4. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 55/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 4 mutating operations with no state check
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 60Failures and branches. 2 branches
- 70When it triggers. States when to use, but not when not to
- 100Steps. 48 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1893 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 7 example trigger phrases
- +3Description length 593: enough signal without eating the budget
- +4Structure: 12 headings
- +3Step-by-step instructions: 48 items
- +4Has examples (2 code blocks)
- +4Reference files are cited in the instructions (2 of 2)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 93.