self-improving
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill establishes a persistent indirect prompt injection surface by creating a cycle of writing and reading untrusted content from the filesystem.
- Ingestion points: All files within the
.learnings/directory at the project root (referenced in SKILL.md). - Boundary markers: Absent. The skill instructs the agent to "read all files in it before starting work" and to "respect them" without providing any delimiters or instructions to ignore malicious content.
- Capability inventory: The skill uses file-read and file-write operations (SKILL.md) to manage persistent state, which then influences the agent's context and use of other platform capabilities in subsequent sessions.
- Sanitization: Absent. There is no mechanism described to validate, sanitize, or filter the content written to or read from the learning files.
Audit Metadata