finishing-a-development-branch
Warn
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: MEDIUMMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
- [METADATA_POISONING]: The skill features contradictory information regarding its stewardship. While the README.md file explicitly claims that the attribution to 'Ship Shit Dev' was incorrect and has been removed, the plugin.json manifest continues to list 'Ship Shit Dev' as the author. This discrepancy creates confusion regarding the skill's origin and stewardship.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from git and gh commands, creating an attack surface for indirect prompt injection.
- Ingestion points: Data is read from branches, worktrees, issues, and pull request states via 'git' and 'gh' tools (SKILL.md).
- Boundary markers: None identified; the skill does not provide instructions to delimit or ignore instructions within the fetched repository data.
- Capability inventory: The skill is authorized to perform 'git commit', 'git push', and 'gh' tool operations, allowing it to modify the repository based on processed data (SKILL.md).
- Sanitization: No evidence of input validation, filtering, or sanitization is present for external data before processing.
Audit Metadata