llm-wiki-setup

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONPERSISTENCEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes Python scripts (init_vault.py, lint-vault.py) designed to manage the wiki vault. These scripts perform operations such as creating directories, copying template files, and setting file execution permissions via os.chmod. These are standard functionalities for the skill's stated purpose of vault management.
  • [PERSISTENCE]: The skill automates the configuration of a local git pre-commit hook using git config core.hooksPath .githooks. This ensures that the provided linting logic is executed automatically whenever a commit is made within the managed vault directory, serving as a productivity and data integrity feature.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted content (e.g., investment reports and meeting transcripts) from the raw/ directory. This creates a potential surface for indirect prompt injection as the agent ingests this data to update the wiki. The skill incorporates security best practices by using yaml.safe_load for parsing metadata and encouraging a human-in-the-loop (HITL) process for data ingestion.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 03:06 AM
Security Audit — agent-trust-hub — llm-wiki-setup