Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md - moving detail out of `SKILL.md` into `references/`
Security audit
Skill Builder
Security checks for vulnerabilities and agentic risk
Overview
This skill is a coherent Markdown-only guide for deciding, creating, or auditing OpenClaw skills, with no hidden code execution or credential handling.
Install this if you want help designing lean skills. Review any generated skill trigger descriptions so they do not fire too broadly, and approve any actual file edits or new scripts based on your normal workflow.
Vulnerability Patterns
- Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
- Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
- Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
- Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
- Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (12)
Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md - moving detail out of `SKILL.md` into `references/`
Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md - moving detail out of `SKILL.md` into `references/`
Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md - moving detail out of `SKILL.md` into `references/`
Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md - moving detail out of `SKILL.md` into `references/`
Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md - moving detail out of `SKILL.md` into `references/`
Ae1
High
- Category
- analysis-evasion
- Confidence
- 100% confidence
- Finding
Referenced artifact was not completely inspected
- Content
md - moving detail out of `SKILL.md` into `references/`
Anti-Refusal Statement
High
- Category
- Anti-Refusal
- Confidence
- 80% confidence
- Finding
Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.
- Content
md Today's LLMs have strong theory of mind. When you explain *why* a step matters, the model generalizes better than when you just bark orders. Good: "Check for existing files first — overwriting without warning breaks trust with the user and creates undo work." Weak: "ALWAYS check for existing files. NEVER overwrite without asking. This is
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 89% confidence
- Finding
This markdown file falls under the vague-trigger category. The guidance to make descriptions 'pushy enough to catch casual/informal phrasings' and to include adjacent concepts can encourage overly broad activation conditions that overlap with normal user speech, without adding limiting constraints or negative examples.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
Telling authors to make descriptions 'slightly aggressive about when to fire' encourages ambiguous activation conditions by design. In the context of a meta-skill that helps define other skills, this pattern can propagate across many skills, leading to systematic overreach, incorrect delegation, and reduced predictability of agent behavior.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
This guidance explicitly recommends broadening trigger phrases to capture adjacent and informal requests, including cases where the user does not explicitly ask for the named capability. In a skill-selection system, that increases the chance of over-triggering, causing the agent to invoke this skill for common requests outside its intended scope and potentially bypass more appropriate tools or narrower skills.
- Content
Autonomous Decision Making
Medium
- Category
- Excessive Agency
- Confidence
- 75% confidence
- Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
- Content
md Good: "Check for existing files first — overwriting without warning breaks trust with the user and creates undo work." Weak: "ALWAYS check for existing files. NEVER overwrite without asking. This is CRITICAL." If you catch yourself writing ALWAYS, NEVER, or CRITICAL in all caps, pause and
Static analysis
No suspicious patterns detected.