git-merge-expert
Pass
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is configured to execute several system-level commands including
git,gh(GitHub CLI),npm, andpnpm. These are used for operational tasks like merging, tagging, and running project-defined scripts (build and test). - [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and processes untrusted data from Git repositories (conflict markers and PR descriptions). This data could be crafted to manipulate the agent's output or command execution.
- Ingestion points: File contents during conflict resolution and PR metadata fetched via
gh pr view. - Boundary markers: Absent; there are no specific instructions to treat content from these sources as data only or to ignore embedded instructions.
- Capability inventory: The agent can perform significant file system and network operations via
gitandgh, and execute project code vianpm/pnpm. - Sanitization: No sanitization or validation of the ingested code or metadata is performed before the agent processes it.
Audit Metadata