coding-worktree-recovery
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use system and Git commands for environment auditing and state management. Examples include
ps,pgrep,lsof,git status,git diff, andstat. These tools are used to verify process ownership and repository integrity. - [PROMPT_INJECTION]: The skill processes untrusted external data such as Git logs, patches, and diffs, creating a surface for indirect prompt injection.
- Ingestion points: Git repository metadata and file contents are processed throughout the recovery workflow as described in SKILL.md.
- Boundary markers: The instructions mandate independent verification steps and controller acceptance gates to mitigate unintended actions.
- Capability inventory: The agent is expected to execute subprocesses for Git and system utilities to perform its core functions.
- Sanitization: The skill relies on mechanical verification methods, such as hash comparisons and manual review gates, rather than automated input filtering.
Audit Metadata