The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill directs the agent to fetch and analyze content from external, attacker-controllable sources including leaderboards (Toolify, ProductHunt, App Store, etc.) and user reviews. This content is integrated into strategy reports without explicit sanitization or instructions to ignore embedded commands, creating a vulnerability surface where malicious external data could influence the agent's behavior.
Ingestion points: External leaderboard URLs (e.g., toolify.ai API) and third-party review sites defined in SKILL.md.
Boundary markers: Absent; the skill does not define specific delimiters for untrusted external data.
Capability inventory: The agent has the capability to read local files (profile.json) and perform network requests to fetch data.
Sanitization: Not present; the skill instructs the agent to "pull the top 10 entries" and "capture one quote-worthy line" directly from the external source.
[EXTERNAL_DOWNLOADS]: The skill provides instructions to fetch data from a specific API endpoint (https://www.toolify.ai/self-api/v1/top/month-top) and other well-known services. While these are documented as data-gathering steps for market research, they involve the automated retrieval of external content at runtime.
[DATA_EXPOSURE]: The skill instructions require the agent to access project-specific metadata stored in the user's home directory at ~/.smtm/projects/{slug}/profile.json. This access is intended for personalizing business strategies based on existing project state.

money-strategy