app-store-changelog
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local bash script
scripts/collect_release_changes.shto retrieve repository metadata and commit history using thegitCLI. - [INDIRECT_PROMPT_INJECTION]: The skill ingests git commit messages, which are external and potentially untrusted data sources, as a primary input for generating release notes.
- Ingestion points: Git commit messages and file lists collected from the repository in
SKILL.mdstep 1. - Boundary markers: The skill does not use specific delimiters or instructions to tell the agent to ignore potential instructions embedded within the commit messages.
- Capability inventory: The skill allows for reading git logs and file lists; it does not involve network exfiltration or direct file writes.
- Sanitization: There is no evidence of filtering or sanitizing commit message content to prevent the agent from following embedded instructions.
Audit Metadata