The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill defines a customMapFunction parameter that accepts a JavaScript function as a string for data transformation. This pattern allows for the execution of arbitrary code logic within the processing pipeline.- [COMMAND_EXECUTION]: The workflow relies on the agent executing shell commands, including node scripts/run_actor.js and curl. The referenced script run_actor.js is not provided within the skill's file structure, representing an unverified external code dependency.- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Twitter (X), which could contain malicious instructions designed to influence the agent's behavior during classification and scoring. 1. Ingestion points: Twitter profile bios, tweets, and follower data are ingested into the agent context (SKILL.md). 2. Boundary markers: No delimiters or safety instructions are defined to separate external data from agent instructions. 3. Capability inventory: The skill uses shell commands for network access (curl) and file system operations via the referenced run_actor.js script. 4. Sanitization: Scraped content is not explicitly sanitized or validated before being analyzed by the agent.

tracking-startup-founders-on-twitter