The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute various commands using the doctrine CLI (e.g., doctrine slice conformance, doctrine check gate, doctrine review conclude) and corresponding MCP tools. These tools are integral to the project's lifecycle management and audit process.
[INDIRECT_PROMPT_INJECTION]: The skill has a potential attack surface for indirect prompt injection as it processes external project files like design.md, slice-nnn.md, and technical specifications to perform its audit functions. Maliciously crafted content in these files could theoretically influence the agent's synthesis or ledger findings.
Ingestion points: design.md, slice-nnn.md, plan.toml, ADRs, tech specs, and phase verification evidence.
Boundary markers: None explicitly mentioned to distinguish between data and instructions.
Capability inventory: Local command execution via the doctrine CLI and MCP server, and file writing to review ledgers.
Sanitization: No specific sanitization or validation of the input file content is defined.

audit