seed-submit-change

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes git rebase, git commit, git push, and gh pr commands to manage feature branches and pull requests. It follows security best practices by using --force-with-lease with an explicit expected remote SHA to prevent accidental data loss or unauthorized overwrites during force pushes.
  • [DYNAMIC_EXECUTION]: The skill executes a local TypeScript script at skills/seed-change-plan/scripts/change-plan.ts using the bun runtime to validate the state of the repository after rebasing. This involves running code that is external to the primary instruction set but resides within the local project structure.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from the seed-change-plan output or existing GitHub PRs to determine the targetBranch, targetRef, and prBase. While the skill includes multiple manual review checkpoints (user confirmation required before commit, push, and PR submission), the parsing of this external data creates an attack surface where malicious input could attempt to influence the agent's branch management decisions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 05:47 AM
Security Audit — agent-trust-hub — seed-submit-change