BF clawguard
Security auditor for ClawHub skills. Run before installing ANY skill — scans SKILL.md and scripts for prompt injection, data exfiltration, shell injection, permission mismatches, and malicious patterns. Returns a PASS / WARN / FAIL verdict with a full breakdown. Triggers on phrases like "scan this skill", "is this skill safe", "audit skill", "check before installing", or "inspect clawhub skill".
Security auditor for ClawHub skills.
As a process F 42/100 · Will not run — References files that are not bundled: scripts/scan.py
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- The text references files that are not there: add them or drop the references.
- 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 · 6
✓ No critical or high findings
Medium and low: 6
-
low Risky intent
intent-offensive-securityREADME.md:80Offensive-security / dual-use content (legitimate for authorised testing; review intended use)✅ No data exfiltration or reverse shell patterns detected
-
low Risky intent
intent-offensive-securitySKILL.md:77Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)Scans all shell scripts for outbound data patterns — curl/wget to unknown domains, DNS tunneling, reverse shell patterns (bash -i, nc -e, /dev/tcp), and base…ded command execution.
detector -
low Risky intent
intent-offensive-securitySKILL.md:130Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)- **❌ FAIL** — Any critical finding: prompt injection, confirmed exfiltration, reverse shell, or credential theft pattern
detector -
low Risky intent
intent-offensive-securitySKILL.md:139Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| Reverse shell pattern | 🔴 Critical | FAIL |
A further 2 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 3. 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") - warning
missing-refreference to a missing file: scripts/scan.py - note
frontmatter-keyunknown frontmatter key "homepage"
Process rating: all ten parameters 42/100
- 0Tools and files. 1 referenced file(s) missing: scripts/scan.py
- 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
- 30Running it twice. 2 mutating operations with no state check
- 40Consistency. Frontmatter name (clawguard) differs from the folder (claw-guard)
- 60Result and completion. Output format stated, no completion criterion
- 70When it triggers. States when to use, but not when not to
- 100Steps. 11 steps
- 100Execution cost. Instruction body is 1591 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 10 top-level sections: this looks like several domains in one skill
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)
- -220 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +5Description quotes 5 example trigger phrases
- +3Description length 398: enough signal without eating the budget
- +4Structure: 21 headings
- +3Step-by-step instructions: 11 items
- +3Output format is stated explicitly
- +4Has examples (5 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 71.