The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a local Node.js script (tools/clis/github-prospects.js) to perform programmatic enrichment of prospect data gathered from GitHub stargazers and forks.
[EXTERNAL_DOWNLOADS]: The skill facilitates data discovery and enrichment through multiple well-known third-party services, including Apollo, Clay, ZoomInfo, Firecrawl, and Browserbase. These tools are used for their intended purpose of gathering firmographic and contact data from public business records.
[PROMPT_INJECTION]: The skill ingests untrusted data from business websites, social media profiles, and directories to qualify prospects, creating an indirect prompt injection surface.
Ingestion points: External data enters the agent context via browser research and tool outputs from sources like Google Maps, Yelp, and individual business websites across all files (e.g., references/local-prospecting.md, references/saas-prospecting.md).
Boundary markers: The instructions do not specify the use of delimiters or safety wrappers for data extracted from these external sites.
Capability inventory: The skill possesses the capability to write local CSV files and execute network-enabled tool calls.
Sanitization: No explicit sanitization or filtering of the extracted website content is described before the data is processed or output into the final lead sheet.
[SAFE]: The skill includes an extensive 'Compliance Guardrails' section in SKILL.md and a dedicated references/compliance.md file. These documents explicitly prohibit malicious behaviors such as bulk scraping, CAPTCHA bypass, or login wall evasion, and provide detailed guidance on maintaining legal lineage for data collection.

prospecting