The Agent Skills Directory

[SAFE]: The skill functions as a requirement auditor. It instructions the agent to read local project documentation to verify completeness and clarity. No malicious behavior or high-risk command execution was detected.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external files, which creates a potential surface for indirect prompt injection.
Ingestion points: The skill reads files located at openspec/changes/<id>/proposal.md, design.md, tasks.md, and openspec/project.md.
Boundary markers: No specific delimiters or boundary markers are defined to isolate untrusted file content from the agent's instructions.
Capability inventory: Based on the instructions, the skill is intended to be read-only and lacks instructions for writing files, executing system commands, or performing network operations.
Sanitization: No sanitization or validation of the ingested markdown content is performed before analysis.

sdd-checklist