BF dcl-skill-auditor
Scan any ClawHub skill before installing it. 534 out of 3,984 ClawHub skills contained critical vulnerabilities — credential theft, prompt injection, data exfiltration. Snyk Research, 2026. DCL Skill Auditor analyzes SKILL.md, scripts, and manifests against 30+ known attack patterns, optionally backed by a real, paid baseline-safety check via the live DCL Trust Oracle MCP server (x402, USDC on Base), and returns a structured PASS / WARN / BLOCK verdict with a cryptographic audit proof. Use this skill before every new install, on skill updates, or in any agent pipeline that requires a pre-execution security checkpoint. Part of the Leibniz Layer™ security suite by Fronesis Labs alongside DCL Policy Enforcer, DCL Prompt Firewall, and DCL Semantic Drift Guard.
As a process F 51/100 · Will not run — References files that are not bundled: scripts/run.sh:17
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 text contains phrases like "ignore previous instructions" or "you are now…". That is an attempt to hijack the agent: it may break your rules, the system limits or company policy.
An honest skill does not need them: state the role and the rules directly without overriding other instructions. Otherwise catalog scanners and corporate filters will block the listing.
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 · 12
✓ No critical or high findings
Medium and low: 12
-
medium Instruction override
en-ignore-previousSKILL.md:57Instruction-override phrase ("ignore previous instructions") (documentation of a security skill)- Instructions to ignore or override system prompts
security skill -
low Risky intent
intent-offensive-securitySKILL.md:64Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- Reverse shell signatures (`/dev/tcp`, `nc -e`, `bash -i`)
-
low Risky intent
intent-offensive-securitySKILL.md:207Offensive-security / dual-use content (legitimate for authorised testing; review intended use)- [ ] Reverse shell: `/dev/tcp/`, `nc -e /bin/bash`, `bash -i >&`
-
low Secrets in code
secret-high-entropy-tokenSKILL.md:276High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)"dcl_proof": "DCL-…a0e",
quoted -
low Secrets in code
secret-high-entropy-tokenSKILL.md:295High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)"dcl_proof": "DCL-…c05",
quoted
A further 7 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 2. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
missing-refreference to a missing file: scripts/run.sh:17
Process rating: all ten parameters 51/100
- 0Tools and files. 1 referenced file(s) missing: scripts/run.sh:17
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 2 mutating operations with no state check
- 60Result and completion. Output format stated, no completion criterion
- 100Steps. 63 steps
- 100Failures and branches. 4 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3367 tokens
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 11 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
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +1No license
- +2Single-language instructions
- +3Description length 766: enough signal without eating the budget
- +4Structure: 31 headings
- +3Step-by-step instructions: 63 items
- +3Output format is stated explicitly
- +4Has examples (10 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 81.