x-twitter-research
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill communicates with an external service at twitr.sh to process data extraction requests and check for completion status. This involves transmitting user-provided search queries or account identifiers to a third-party domain.
- Evidence: The skill explicitly uses 'https://twitr.sh/api/tools/x_extract' and 'https://twitr.sh/api/snapshots/' for its core functionality.
- [PROMPT_INJECTION]: The skill instructs the agent to fetch and summarize data from an external URL, which represents an indirect prompt injection surface where malicious data could influence agent behavior.
- Ingestion points: Data is ingested from the 'download_url' provided in the job completion response.
- Boundary markers: There are no instructions providing delimiters or 'ignore embedded instructions' warnings for the downloaded content.
- Capability inventory: The agent is authorized to fetch external files and summarize their contents into the conversation context.
- Sanitization: No sanitization or validation of the downloaded dataset's content is described or implemented in the skill instructions.
Audit Metadata