github-issues-task-manager

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the cloning of the oglofus/skills repository from GitHub into a temporary directory as part of its installation procedure.
  • [REMOTE_CODE_EXECUTION]: The agent is instructed to execute an install.sh script located within the cloned vendor repository.
  • [COMMAND_EXECUTION]: Makes extensive use of the gh tool and git to perform repository metadata retrieval, issue tracking, and PR management. It uses jq to safely construct JSON payloads for GitHub API interactions.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources such as GitHub issue comments and pull request diffs, which serves as a surface for indirect prompt injection.
  • Ingestion points: gh issue view --comments and gh pr diff in SKILL.md.
  • Boundary markers: None explicitly defined for isolating external content during the reading phase.
  • Capability inventory: The skill has access to gh api, git commit, git push, and arbitrary shell command execution through the gh tool.
  • Sanitization: Employs jq for output formatting to the API, but lacks sanitization or boundary enforcement for input read from diffs or issue discussions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 07:31 AM
Security Audit — agent-trust-hub — github-issues-task-manager