triage-pr

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it fetches and processes untrusted pull request comments which could influence the agent's behavior during categorization and file editing.
  • Ingestion points: PR metadata, inline comments, and GraphQL review threads fetched via the gh tool in Step 2.
  • Boundary markers: The skill lacks explicit delimiters or instructions to treat external comment text as untrusted data.
  • Capability inventory: The skill has access to powerful tools including Bash, Read, Edit, and Write, which could be abused if the agent follows instructions embedded in a PR comment.
  • Sanitization: There is no evidence of sanitization or filtering of the fetched PR comment content before it is passed to the agent for processing.
  • [COMMAND_EXECUTION]: The skill performs shell command interpolation by using the $ARGUMENTS variable directly within a bash command in Step 1 (gh pr view <num>). This presents a potential command injection risk if a user provides an argument containing shell metacharacters like semicolons or backticks.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 06:18 PM
Security Audit — agent-trust-hub — triage-pr