The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests raw tweet text and metadata from external X/Twitter accounts, which are untrusted sources. This data is used to generate research reports and content recommendations.
Ingestion points: Tweet content is fetched from the Apify API via scripts/fetch_tweets.py and stored in raw.json.
Boundary markers: The workflow does not explicitly define delimiters or instructions for the agent to ignore potentially malicious instructions embedded within the fetched tweet text during report generation.
Capability inventory: The skill possesses the ability to write files to the local file system (report.md, outliers.json) and execute Python scripts via shell commands.
Sanitization: No sanitization or filtering of the tweet text is performed before it is presented to the agent for analysis.
[EXTERNAL_DOWNLOADS]: The skill connects to Apify's well-known automation platform to fetch tweet data using the apify-client library.
Evidence: scripts/fetch_tweets.py invokes the apidojo/tweet-scraper Actor on Apify to retrieve search results. This is a standard integration with a well-known technology service.

x-research