BD secureclaw
Security skill for OpenClaw agents (7-framework aligned). 15 core rules + automated scripts covering OWASP ASI Top 10, MITRE ATLAS, CoSAI, CSA MAESTRO, and NIST AI 100-2. Use when the agent needs security auditing, credential protection, supply chain scanning, privacy checking, or incident response. By Adversa AI (https://adversa.ai). v2.2.0.
Security skill for OpenClaw agents (7-framework aligned).
As a process D 46/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions
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
- 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 · 9
✓ No critical or high findings
Medium and low: 9
-
medium Dangerous commands
cmd-pipe-to-shellSKILL.md:33Downloads and executes remote code from an unrecognised host (pipe to shell) (documentation of a security skill)2. Before executing destructive or sensitive commands (rm -rf, curl|sh,
security skill -
low Risky intent
intent-offensive-securityREADME.md:3Offensive-security / dual-use content (legitimate for authorised testing; review intended use)Developed by [Adversa AI](https://adversa.ai) — Agentic AI Security and Red Teaming Pioneers.
-
low Risky intent
intent-offensive-securitySKILL.md:13Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)Rule 3: MAESTRO L4 (Infrastructure) | NIST: Privacy (credential harvesting)
detector -
low Dangerous commands
cmd-privilegeSKILL.md:34Privilege escalation / world-writable permissions (documentation of a security skill)eval/exec, chmod 777, credential access, mass email/message sends,
security skill
A further 5 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 17. 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
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 8 mutating operations with no state check
- 40Consistency. Frontmatter name (secureclaw) differs from the folder (secureclaw-skill)
- 55Failures and branches. 1 branches
- 60Tools and files. Uses tools (web, git) that frontmatter does not declare
- 100Steps. 20 steps
- 100Execution cost. Instruction body is 1431 tokens
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
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
- -33 of 9 scripts are never mentioned in SKILL.md
- +1No license
- +2Single-language instructions
- +3Description length 344: enough signal without eating the budget
- +4Structure: 3 headings
- +3Step-by-step instructions: 20 items
- +4Has examples (5 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 81.