The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The file lib/api.ts programmatically reads sensitive file paths, specifically ~/.env.keys and ~/.config/env/global.env, to extract the X_BEARER_TOKEN for API authentication. While this is the documented method for the tool to function, it involves accessing a global secret store.
[PROMPT_INJECTION]: The skill ingests untrusted data from the X API and syndication endpoints, creating an indirect prompt injection surface.
Ingestion points: Data is fetched from api.x.com (search, profiles, threads) and cdn.syndication.twimg.com (syndication endpoint) in lib/api.ts and lib/syndication.ts.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the SKILL.md or the script logic when presenting tweet content to the agent.
Capability inventory: The skill can perform network requests to the X API and write configuration and research reports to the local file system.
Sanitization: The tool performs basic cleaning of text in lib/format.ts but does not sanitize for malicious instructions within fetched tweet content.

xray