Changelog Curator
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it requires the agent to parse and categorize untrusted data from external sources.
- Ingestion points: The skill ingests raw
git logoutput, pull request titles, and existingCHANGELOG.mdfile content (SKILL.md). - Boundary markers: Absent. There are no instructions for the agent to use delimiters or ignore embedded instructions within the git log or PR data.
- Capability inventory: The agent has file-system write capabilities to update or create the
CHANGELOG.mdfile. - Sanitization: Absent. The instructions do not specify any validation or sanitization of the input strings before they are processed and written to the documentation.
Audit Metadata