The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute various ontoly CLI commands such as ontoly build ., ontoly coverage ., and ontoly query. These commands are integral to the skill's primary purpose of analyzing software repositories for onboarding.
[EXTERNAL_DOWNLOADS]: The skill references external documentation and an installation service at ontoly.xyz and skills.sh. These links are provided for informational and setup purposes for the Ontoly toolset.
[INDIRECT_PROMPT_INJECTION]: The skill processes external repository files and graph data, creating a potential surface for indirect prompt injection.
Ingestion points: The agent reads from .ontoly/SoftwareGraph.json and performs fallback inspections of repository source files as specified in SKILL.md and reference/workflow.md.
Boundary markers: The instructions mandate a 'graph-first' approach, require specific evidence citations (node IDs, graph hashes), and define confidence levels to mitigate reliance on potentially tainted source data, as documented in reference/best-practices.md.
Capability inventory: The skill utilizes various analysis capabilities including ExplainArchitecture, FindEntrypoints, and file inspection tools via the Ontoly MCP (documented in reference/mcp.md).
Sanitization: The skill does not explicitly describe sanitization or escaping of the content read from repository files, relying instead on the agent's internal safety filters and the structured nature of the graph evidence.

codebase-onboarding