The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data fetched from external Twitter/X profiles, including tweet text and replies. This data is ingested into the agent's context without explicit boundary markers or sanitization, creating a surface for indirect prompt injection attacks where malicious instructions hidden in social media posts could influence the agent's behavior.
Ingestion points: External Twitter/X profile data fetched via the Apify API.
Boundary markers: None present in the skill instructions to separate untrusted data from agent commands.
Capability inventory: The skill is designed for data collection, formatting (CSV/JSON), and potential downstream analysis using local shell commands.
Sanitization: No evidence of sanitization or filtering of the fetched content is mentioned in the skill definition.
[DYNAMIC_EXECUTION]: The skill supports a customMapFunction input parameter, which allows for the execution of arbitrary JavaScript code on the Apify platform to transform output data. While this is a standard feature of the integrated service, it represents a dynamic execution vector if the logic is influenced by untrusted sources.
[COMMAND_EXECUTION]: The skill documentation provides examples for executing shell commands using node scripts/run_actor.js and curl to interact with the Apify API, which is the intended mechanism for the skill's functionality.

scraping-twitter-profile-tweets