skills/yeachan-heo/oh-my-codex/wiki/Gen Agent Trust Hub

wiki

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes omx wiki commands (e.g., wiki_ingest, wiki_query, wiki_lint) to perform CRUD operations on wiki files. These are project-scoped utility commands intended for knowledge management.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as an interface for reading and searching markdown files within the repository (omx_wiki/*.md). While these files are local project data, they represent a surface where untrusted content could potentially influence the agent's behavior during a query or read operation.
  • Ingestion points: Files located in omx_wiki/*.md, omx_wiki/index.md, and omx_wiki/log.md.
  • Boundary markers: None explicitly defined in the instructions for separating content from instructions.
  • Capability inventory: Commands for reading, searching, adding, and deleting files within the project scope.
  • Sanitization: No specific sanitization or filtering of the wiki content is mentioned before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 04:53 PM
Security Audit — agent-trust-hub — wiki