The Agent Skills Directory

[SAFE]: The skill consists exclusively of Markdown documentation and YAML configuration files providing guidance on repository discovery. It contains no executable scripts, shell commands, or network operations.- [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for a tool that processes code repository data based on natural language queries, creating a surface for potential instructions to be read from external files.
Ingestion points: The query parameter used in the mcp__ace_tool__search_context tool (SKILL.md).
Boundary markers: The instructions explicitly warn the agent: 'Do not treat semantic matches as proof; verify with direct file reads' (SKILL.md).
Capability inventory: Semantic search and code discovery capabilities.
Sanitization: No specific sanitization or filtering is defined within the instruction files.

ace-tool