The Agent Skills Directory

[SAFE]: The skill is designed for strategic business analysis and uses a multi-agent system to ensure logical consistency. No malicious code or behavioral patterns were detected during the audit.
[COMMAND_EXECUTION]: The skill includes utility scripts written for the Bun runtime. These scripts, such as manifest-sync.ts and scaffold-eval-loop.ts, automate the management of the skill's local environment, including directory scaffolding and artifact indexing. These operations are restricted to the project's local file system.
[DATA_EXFILTRATION]: The external-check-agent uses WebSearch to investigate market trends and competitor activity. This network access is aligned with the skill's stated purpose and does not involve the transmission of sensitive local data or credentials.
[PROMPT_INJECTION]: The instructions provided to the agents are strictly procedural, focused on maintaining the MECE logic framework and the If/Then/Because hypothesis structure. No attempts to override safety filters or redirect agent behavior were found.
[SAFE]: The skill implements a critic-agent with a 10-point quality gate, which serves as a security and quality control mechanism to ensure that the final output is grounded in evidence and adheres to the structural requirements of the diagnosis.

diagnose