analyzing-twitter-sentiment-for-topic

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes public tweets, which constitute untrusted external data. This creates a surface where malicious content within tweets could attempt to influence the agent's sentiment classification or summary generation.
  • Ingestion points: Data retrieved from the apidojo~tweet-scraper Apify actor.
  • Boundary markers: Absent. The instructions do not specify the use of delimiters or clear separation between the data (tweet text) and the classification tasks.
  • Capability inventory: The skill utilizes shell commands (node, curl) and network access to interact with the Apify API.
  • Sanitization: Absent. There is no evidence of content sanitization or validation of the retrieved tweet text before it is analyzed by the agent.
  • [COMMAND_EXECUTION]: The skill instructs the user and agent to execute local shell commands using node and curl to run scripts and interact with the Apify REST API. While these are intended for the skill's primary function, they involve direct command execution in the host environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 06:40 PM
Security Audit — agent-trust-hub — analyzing-twitter-sentiment-for-topic