skills/trailofbits/skills/pr-improver/Gen Agent Trust Hub

pr-improver

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to run git commands, such as git rev-parse and git diff, to identify the project root and the set of changed files requiring improvement.- [DYNAMIC_EXECUTION]: The skill dynamically resolves the location of its core execution script (workflows/improve.js) by searching predefined plugin directories and the current working directory, then executes the resolved path using the Workflow tool.- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local environment which could be exploited to influence agent behavior.
  • Ingestion points: Ingests repository file names and diff content via git diff and processes the results of an external reviewer skill from the TaskOutput tool.
  • Boundary markers: No explicit instructions for boundary markers or instruction-ignoring delimiters are provided in the skill body.
  • Capability inventory: The skill has the ability to execute shell commands, read files, and trigger additional workflows.
  • Sanitization: There is no evidence of sanitization or validation performed on the ingested diff content before it is passed to the workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 02:03 AM
Security Audit — agent-trust-hub — pr-improver