The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local CLI command roundfix spec check <slug> --stage techspec to validate the generated artifacts. The slug parameter is sourced from user-supplied arguments.
[INDIRECT_PROMPT_INJECTION]: The skill processes various external project documents as reference material, which creates an attack surface where these files could potentially influence agent behavior.
Ingestion points: Data is ingested from $ARGUMENTS and local project files including docs/specs/<slug>/_prd.md, docs/adr/, CONTEXT.md, and architectural guides in docs/agents/.
Boundary markers: The skill does not define specific delimiters to isolate external content from the agent's core instructions.
Capability inventory: The skill has access to file writing operations and the ability to execute the roundfix subprocess.
Sanitization: No sanitization or content filtering is performed on the ingested documentation data.

write-techspec