The Agent Skills Directory

[SAFE]: The skill acts as a static documentation repository for the BEO system. It possesses no writable surfaces and performs no network operations.
[COMMAND_EXECUTION]: The skill documents the syntax for the 'br' and 'bv' CLI tools, which are used by the broader system for task management and state synchronization. This skill only provides references and does not execute these commands directly.
[PROMPT_INJECTION]: The skill defines a surface for indirect prompt injection (Category 8) because it is designed to retrieve data from local artifacts in the '.beads/' directory. 1. Ingestion points: Reads files like CONTEXT.md and PLAN.md from '.beads/artifacts/' as described in the artifacts reference. 2. Boundary markers: Data is structured using Markdown headers and JSON schemas, providing organizational separation between metadata and content. 3. Capability inventory: The skill documents CLI execution and file writing, which are capabilities available to other skills in the BEO suite (e.g., 'beo-execute'). 4. Sanitization: No specific mechanisms are mentioned for sanitizing the content of ingested artifacts before they are presented to the agent.

beo-reference