gentle-ai-chained-pr

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill analyzes existing pull request content to guide its splitting logic, which constitutes a data ingestion surface.\n
  • Ingestion points: The skill retrieves pull request metadata and descriptions using the gh pr view command as referenced in references/chaining-details.md.\n
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to disregard potential instructions embedded within the ingested PR content.\n
  • Capability inventory: The skill utilizes gh pr create to generate new pull request branches and bodies as detailed in references/chaining-details.md.\n
  • Sanitization: No specific sanitization or validation logic is defined for the content extracted from pull requests, though the agent's actions are restricted to PR management tasks.\n- [COMMAND_EXECUTION]: The skill executes shell commands using the official GitHub CLI to perform its functions.\n
  • Evidence: The skill defines the use of gh pr view and gh pr create within its workflow in references/chaining-details.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 03:31 PM
Security Audit — agent-trust-hub — gentle-ai-chained-pr