sdlc-setup-wiki

Pass

Audited by Gen Agent Trust Hub on Nov 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a Python utility (wiki_llm.py) to execute Git commands via the subprocess.run function. These calls are implemented using argument lists rather than shell strings, and path arguments are properly escaped using shlex.quote in relevant sections (e.g., in wikilib/link.py). This implementation minimizes the risk of command injection.
  • [DYNAMIC_EXECUTION]: The validation component (wikilib/validate.py) utilizes importlib.import_module to load rule modules dynamically from the local wikilib/rules/ directory. While dynamic loading is present, it is restricted to the internal package structure and used to facilitate an extensible validation framework rather than executing external or untrusted code.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes Markdown and YAML frontmatter from the project workspace to build and validate the knowledge bundle. This creates a surface for indirect prompt injection if an attacker provides malicious content in the wiki files. However, the skill treats this data as structured input for its validation and synchronization logic, and the agent's instructions require user approval for all changes, which serves as a mitigation layer.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 13, 2026, 09:20 AM