wiki-commit

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes standard shell commands for repository management, including git remote, git fetch, git pull, git status, and git push. These operations are consistent with the skill's primary function of synchronizing a wiki repository.\n- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing untrusted repository data to automate decision-making.\n
  • Ingestion points: Reads the contents of general documentation files and wiki-manifest.yaml (SKILL.md) to generate semantic commit messages and resolve merge conflicts.\n
  • Boundary markers: No explicit delimiters are used to separate repository data from the agent's instructions, nor are there warnings to ignore instructions embedded in the data.\n
  • Capability inventory: The skill can execute shell commands (git) and modify repository files (SKILL.md).\n
  • Sanitization: The skill lacks explicit sanitization or validation logic for the content of the files it interprets. Note that the workflow includes a mandatory human-in-the-loop confirmation step for commits and pushes, which serves as a significant mitigation against automated exploitation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 10:37 AM