AD safe-skill
Programmatic security scanner for AI agent skills. Performs AST analysis, regex pattern matching, Shannon entropy detection, URL/IP extraction, and permission scope inference — with Markdown context awareness to reduce false positives. Use before installing any skill from ClawHub, GitHub, or untrusted sources. Trigger: "scan this skill", "vet this skill", "is this skill safe", "check skill security", "审查这个skill", "扫描skill安全性", "这个skill安全吗".
As a process D 46/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions
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 · 10
✓ No critical or high findings
Medium and low: 10
-
low Risky intent
intent-offensive-securityioc/c2-ips.txt:13Offensive-security / dual-use content (legitimate for authorised testing; review intended use)# Reverse shell endpoint
-
low Risky intent
intent-offensive-securityioc/c2-ips.txt:14Offensive-security / dual-use content (legitimate for authorised testing; review intended use)54.91.154.110|clawhavoc-revshell|2026-01-28|Reverse shell target port 13338
-
low Risky intent
intent-offensive-securityioc/malicious-domains.txt:15Offensive-security / dual-use content (legitimate for authorised testing; review intended use)burpcollaborator.net|exfil|generic|Pentest tool (suspicious in skills)
-
low Risky intent
intent-offensive-securityioc/malicious-skill-patterns.txt:47Offensive-security / dual-use content (legitimate for authorised testing; review intended use)^polymarket-all-in-one$|reverse-shell|Contains reverse shell backdoor (noreplyboter)
-
low Dangerous commands
cmd-pipe-to-shellREADME.md:81Downloads and executes remote code from an unrecognised host (pipe to shell) (documentation table row; documentation of a security skill)| `rce` | `curl\|bash`, `eval()`, `exec()`, `subprocess`, `os.system` | 12 |
tablesecurity skill -
low Dangerous commands
cmd-pipe-to-shellREADME.md:108Downloads and executes remote code from an unrecognised host (pipe to shell) (documentation table row; documentation of a security skill)| Critical | 25 | `curl\|bash`, `~/.ssh` access, crontab injection |
tablesecurity skill -
low Risky intent
intent-offensive-securitySKILL.md:59Offensive-security / dual-use content (legitimate for authorised testing; review intended use)obfuscation, privilege escalation, agent identity theft, persistence/backdoor, suspicious
A further 3 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 12. 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 46/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (web, python) that frontmatter does not declare
- 100Steps. 9 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 964 tokens
- 100Running it twice. No mutating operations
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 444: enough signal without eating the budget
- +4Structure: 14 headings
- +3Step-by-step instructions: 9 items
- +4Has examples (2 code blocks)
- +3All 2 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 92.