writing-changelogs
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to summarize work performed, which involves processing external data such as git commit messages, file diffs, and project history. This presents a surface for indirect prompt injection if the source data contains malicious instructions designed to manipulate the generated summary.
- Ingestion points: Data enters the context from the repository history and completed tasks as described in SKILL.md.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands in the source data are provided.
- Capability inventory: The skill is restricted to text generation and summarization; it does not invoke network operations or sensitive file writes.
- Sanitization: There are no instructions for sanitizing or escaping content retrieved from external logs or commit messages.
Audit Metadata