The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill includes several TypeScript scripts (fetch_blog.ts, fetch_x.ts, fetch_nostr.ts) that use the fetch API and WebSockets to retrieve data from external domains. These include lynalden.com, twitter.com, archive.org, and public Nostr relays (primal.net, damus.io, nos.lol). These connections are necessary for the skill's primary purpose of providing up-to-date analysis.
[COMMAND_EXECUTION]: The SKILL.md file instructs the agent to run local TypeScript scripts using the bun runtime (e.g., bun .agents/skills/investor-lyn-alden/scripts/current_context.ts). These scripts are provided with the skill and are used to aggregate data for the agent's context.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external social media feeds and RSS feeds.
Ingestion points: External data is fetched via scripts/fetch_blog.ts, scripts/fetch_x.ts, and scripts/fetch_nostr.ts and returned to the agent.
Boundary markers: The orchestrator script (current_context.ts) organizes fetched content under clear Markdown headers (e.g., '## X (@LynAldenContact)').
Capability inventory: The agent uses this data to answer user questions and synthesize conclusions. The skill does not grant the agent high-privilege write access or execution capabilities based on this data.
Sanitization: The scripts perform basic cleaning, such as stripping HTML tags and decoding entities, to treat the fetched content as raw text rather than executable instructions.

investor-lyn-alden