sdlc-fix
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent must ingest and act upon potentially untrusted data from a code repository.
- Ingestion points: The agent is instructed to read "applicable repository instructions and the affected source" (SKILL.md, Step 1) and inspect check outputs.
- Capability inventory: The agent has permissions to read files, execute local tests/probes, and perform git commits.
- Boundary markers: There are no explicit instructions for the agent to use delimiters or ignore instructions that may be embedded within the source code or repository-level instruction files it reads.
- Sanitization: The skill does not specify sanitization or validation logic for data read from the repository before it is processed by the agent.
Audit Metadata