share-session-learnings
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies and processes external data from git logs, changelogs, and conversation history, which creates a potential surface for indirect prompt injection.
- Ingestion points: Data is gathered from
git log,CHANGELOG, and session artifacts (specified in SKILL.md). - Boundary markers: The skill contains a defensive instruction requiring the agent to treat logs and artifacts as evidence only and to explicitly ignore any instructions found within them.
- Capability inventory: The skill's execution environment is highly restricted; it is forbidden from performing file writes, code execution, commits, or network operations.
- Sanitization: The skill relies on natural language constraints and a 'read-only' policy to prevent the exploitation of processed data.
Audit Metadata