The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from existing project files and other agent skill files when operating in 'Reverse Mode' to document existing systems.
Ingestion points: The skill reads content from docs/context/system-reverse-spec.md, brownfield-map.md, graph-spec.md, and ns-langgraph-agents/SKILL.md (SKILL.md, Reverse mode section).
Boundary markers: The instructions do not specify delimiters or provide warnings to ignore instructions that might be embedded within the project files it processes.
Capability inventory: The skill has the ability to read arbitrary project files and write architecture decision records to docs/specs/agent-architecture.md (Step 8).
Sanitization: No explicit sanitization, validation, or filtering of the content ingested from external project files is required before it is used to generate report output.
[COMMAND_EXECUTION]: The skill instructions direct the agent to use command-line or filesystem utilities to locate specific skill files.
Evidence: The 'LangGraph peer' section in SKILL.md specifies using grep or Glob to search for ns-langgraph-agents/SKILL.md within the .agents/skills/ or skills/ directories.

ns-multi-agent-architect