BC skill-guard
SkillShield v4 — Ultimate security scanner for OpenClaw skills. 65 security checks, SARIF v2.1.0 output, CI/CD ready, campaign detection, C2 IP blocklist, known malicious actor database, macOS attack detection, agent config tampering, LLM tool exploitation, pre-commit hooks, and more. Python 3 stdlib only. Single file. Zero dependencies.
As a process C 55/100 · Has gaps — weak spots: when it triggers, failures and branches, consistency
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- 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 · 12
✓ No critical or high findings
Medium and low: 12
-
low Dangerous commands
cmd-background-processscripts/skillguard.py:873Starts a background / autostarted process (string literal in code, not executed; documentation of a security skill)recommendation="disown + network = detached backdoor process.",
code literalsecurity skill -
low Risky intent
intent-offensive-securitySKILL.md:204Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| SS-016 | Reverse shell pattern | CRITICAL | 10 |
-
low Dangerous commands
cmd-background-processSKILL.md:251Starts a background / autostarted process (documentation of a security skill)| SS-063 | Nohup/disown + network | CRITICAL | 9 |
security skill -
low Risky intent
intent-offensive-securitySKILL.md:265Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- `91.92.242.30` — AMOS C2 server
-
low Risky intent
intent-offensive-securitySKILL.md:266Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- `54.91.154.110` — AMOS C2 server
A further 7 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 4. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-no-whendescription does not say WHEN to use the skill (no "use when")
Process rating: all ten parameters 55/100
- 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
- 30Running it twice. 5 mutating operations with no state check
- 40Consistency. Frontmatter name (skill-guard) differs from the folder (skillshield)
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Steps. 16 steps
- 100Execution cost. Instruction body is 2432 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low The response is described with custom markup (4 tags): a typed call is more reliable
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
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- -283 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +3Description length 339: enough signal without eating the budget
- +4Structure: 19 headings
- +3Step-by-step instructions: 16 items
- +3Output format is stated explicitly
- +4Has examples (9 code blocks)
- +3All 1 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 76.