The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill includes an audit feature ('/fxmind audit') that processes user-provided Lua scripts, which could contain malicious instructions designed to influence the agent. * Ingestion points: The agent is instructed in audit-passes.md (Pass 0) to read every Lua file defined in the project's manifest. * Boundary markers: The skill does not provide delimiters or specific instructions for the agent to treat file content as untrusted data. * Capability inventory: The agent has capabilities to fetch external documentation and generate implementation code. * Sanitization: No sanitization process is described for handling the ingested script content.
[EXTERNAL_DOWNLOADS]: The skill references several external domains and repositories for developmental support. * Evidence: References include docs.fivem.net, overextended.dev, and forge.plebmasters.de, which are established community and documentation sites for FiveM. It also references author-owned GitHub repositories (proelias7) for development utilities.

fivem-development