The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes several MCP tools (mcp__graph-brain__search-knowledge, mcp__graph-brain__get-knowledge, mcp__graph-brain__search-by-tags) to search, retrieve, and update content in a knowledge graph. It also performs file system writes by appending to .brain/activity-log.json.
[PROMPT_INJECTION]: Indirect prompt injection surface detected.
Ingestion points: Note content and changelog data are retrieved from the knowledge base using mcp__graph-brain__get-knowledge in SKILL.md.
Boundary markers: The instructions do not specify any boundary markers or instructions to ignore embedded commands within the processed data.
Capability inventory: The skill possesses the capability to read/write to the knowledge base and append to local files.
Sanitization: There is no evidence of sanitization, filtering, or validation of the content retrieved from external notes before it is processed or displayed.

brain-history