context-handoff
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests data from various repository files to generate summaries.
- Ingestion points: AGENTS.md, git status/diff output, decision logs, wiki pages, and test results from the local filesystem.
- Boundary markers: Absent in the template.
- Capability inventory: The skill is capable of reading and writing files and executing git commands.
- Sanitization: The instructions explicitly advise the agent not to copy secrets or full logs and to treat fresh command output as authoritative over stored text.
- [SAFE]: The primary purpose of the skill is legitimate project state management. It follows security best practices by including defensive instructions against credential exposure and verifying state through authoritative repository commands.
Audit Metadata