The Agent Skills Directory

[SAFE]: The skill's instructions are focused entirely on product management methodologies. It provides a structured framework for analyzing features and translating them into business and customer impacts without any suspicious command execution or data exfiltration.
[EXTERNAL_DOWNLOADS]: The skill includes links to informational articles on productcompass.pm. These are provided as educational resources for the user and are from a well-known industry publication, posing no security risk.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided roadmaps and strategic documents. 1. Ingestion points: User input and external documents are ingested in SKILL.md for analysis. 2. Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present. 3. Capability inventory: The skill utilizes web search for context and the ability to save the final roadmap as a markdown file. 4. Sanitization: No specific text sanitization is described. This is an expected attack surface for a skill whose primary purpose is document processing and does not escalate the verdict.

outcome-roadmap