The Agent Skills Directory

[SAFE]: The skill instructions and logic demonstrate no malicious intent, such as prompt injection, credential harvesting, or unauthorized data exfiltration. The skill follows established architectural patterns for project data management.
[COMMAND_EXECUTION]: The skill executes local maintenance scripts (e.g., 'cascade-rename.sh', 'validate-entities.sh', and 'project-status.sh') located within the plugin's root directory. These operations are restricted to maintaining the integrity and structure of the user's project portfolio.
[EXTERNAL_DOWNLOADS]: The skill invokes a specialized subagent ('cogni-portfolio:market-researcher') to perform market research tasks. This is a vendor-provided capability that aligns with the skill's primary function and does not constitute an untrusted remote dependency.

markets