AD carapace
Runtime security hardening for OpenClaw agents. Protects against prompt injection, data exfiltration, credential leaks, and unauthorized operations. Use when setting up agent security, performing security audits, protecting credentials, preventing data leaks, hardening agent configurations, or defending against indirect prompt injection attacks. Complements pre-installation skill scanners by hardening the agent itself at runtime.
As a process D 48/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions
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 · 6
✓ No critical or high findings
Medium and low: 6
-
low Risky intent
intent-offensive-securityreferences/attack-vectors.md:41Offensive-security / dual-use content (legitimate for authorised testing; review intended use)## 3. Credential Harvesting
-
low Risky intent
intent-offensive-securityreferences/attack-vectors.md:57Offensive-security / dual-use content (legitimate for authorised testing; review intended use)## 4. Privilege Escalation via Config Modification
-
low Risky intent
intent-offensive-securityreferences/audit-template.md:109Offensive-security / dual-use content (legitimate for authorised testing; review intended use)## 4. Penetration Testing
A further 3 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
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 48/100
- 0Result and completion. Does not say what the result is
- 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
- 40Consistency. Frontmatter name (carapace) differs from the folder (eridian)
- 60Tools and files. Uses tools (bash) that frontmatter does not declare
- 65Failures and branches. 3 branches
- 100Steps. 64 steps
- 100Execution cost. Instruction body is 1414 tokens
- 100Running it twice. Mutating operations check current state
- 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
- +1No license
- +2Single-language instructions
- +3Description length 433: enough signal without eating the budget
- +4Structure: 15 headings
- +3Step-by-step instructions: 64 items
- +4Has examples (1 code blocks)
- +4Reference files are cited in the instructions (3 of 3)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 88.