hierarchical
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local shell commands to retrieve repository state and project configuration.
- Evidence: The context section uses shell execution (
!) to rungit branch,git status,git log, andgit diffto provide the agent with information about the current branch and code changes. - Evidence: Uses shell logic to check for the presence of manifest files like
package.json,Cargo.toml, andgo.modto identify available test frameworks. - [PROMPT_INJECTION]: The skill is subject to indirect prompt injection through the codebase it analyzes.
- Ingestion points: Git diffs and log outputs are ingested directly into the prompt context for review by specialized subagents (SKILL.md).
- Boundary markers: Absent. There are no explicit delimiters or instructions provided to the subagents to ignore or isolate instructions found within the code being reviewed.
- Capability inventory: The skill has the ability to execute local git commands and orchestrate multiple specialized agents via the
Tasktool. - Sanitization: None. The raw output of code changes is passed to the reviewers without validation or escaping.
Audit Metadata