Skills
skills/xexr/marketplace/gastown-upstream-sync/Gen Agent Trust Hub

gastown-upstream-sync

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: Indirect prompt injection surface identified in the automated processing of external git metadata.\n
  • Ingestion points: The skill fetches commit messages and pull request titles using git log and gh pr list in SKILL.md.\n
  • Boundary markers: There are no delimiters or instructions provided to the agent to ignore instructions embedded within the commit history or PR descriptions.\n
  • Capability inventory: The skill has significant local capabilities, including binary installation (make install), code execution (gt doctor), and remote repository modification (git push).\n
  • Sanitization: No sanitization or filtering is applied to the fetched metadata before it is presented to the agent or used in decision-making.\n- [COMMAND_EXECUTION]: The skill performs local compilation and tool execution.\n
  • Evidence: Executes make build, make install, and gt doctor --fix to update the local environment and binaries based on the synchronised source code.\n- [EXTERNAL_DOWNLOADS]: Synchronizes source code from a remote repository.\n
  • Evidence: Uses git fetch upstream to retrieve updates from steveyegge/gastown on GitHub.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 10:40 AM