uncomplect
Pass
Audited by Gen Agent Trust Hub on Nov 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill metadata references a public GitHub repository hosted by the author (joelhooks/skills) as its documentation basis. This is a standard reference to a vendor-owned resource.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze source code, tests, and runtime artifacts from the local project environment.
- Ingestion points: Project source files, tests, and
.brain/decision records referenced in SKILL.md. - Boundary markers: None explicitly defined for the agent's ingestion process, though the skill provides instructions for implementing them in the target code.
- Capability inventory: Reading local files and proposing/applying code modifications.
- Sanitization: The skill specifically advocates for security-positive patterns, such as using Effect Schemas to decode and validate untrusted external inputs at application boundaries.
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected. The skill's primary focus is on improving code maintainability and reducing complexity.
Audit Metadata