The Agent Skills Directory

[SAFE] (SAFE): The skill is a documentation-only resource designed to assist an AI agent in performing static code analysis of Rust source code.
[Indirect Prompt Injection] (LOW): The skill is designed to read and process untrusted .rs files provided by the user. While this creates a surface for indirect prompt injection via code comments, the risk is mitigated by the skill's restricted toolset.
Ingestion points: The agent uses Read, Grep, and Glob tools to ingest content from files matching **/*.rs as defined in SKILL.md.
Boundary markers: Absent. The skill does not explicitly instruct the agent to distinguish between its own logic and instructions that might be embedded in the code it analyzes.
Capability inventory: Very limited. The skill only allows file reading and searching; it lacks capabilities for network access, shell execution (exec/eval), or file writing, preventing any significant exfiltration or system compromise.
Sanitization: Absent. The skill treats ingested code as data to be evaluated against the provided safety rules.
[Metadata Poisoning] (SAFE): The skill's metadata accurately reflects its purpose. The use of 'CRITICAL' in the description is a priority marker for the agent's trigger system and is not used to override global safety constraints.

unsafe-checker