recover-branch-context
Pass
Audited by Gen Agent Trust Hub on Feb 18, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [Command Execution] (SAFE): The skill utilizes git commands such as
git log,git diff, andgit branchto extract branch metadata and divergence points. These are standard, non-destructive read operations. - [Indirect Prompt Injection] (SAFE): The skill analyzes potentially untrusted data from commit logs and repository files.
- Ingestion points: Commit messages and file content are read to summarize intent.
- Boundary markers: Absent; the skill relies on the agent's ability to contextualize code versus instructions during analysis.
- Capability inventory: Limited to local git reads and Linear ticket retrieval via an MCP tool.
- Sanitization: None; data is processed as diagnostic input. The risk level is negligible as the operations are restricted to read-only analysis.
Audit Metadata