social
Pass
Audited by Gen Agent Trust Hub on Jul 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes shell command templates in
references/listening.mdthat utilizecurl,jq, andxmllintto fetch data from external JSON and RSS endpoints. These commands are intended to be executed by the agent to retrieve real-time social media content for analysis. - [PROMPT_INJECTION]: The skill implements a social listening and triage loop that ingests untrusted data from the web, which may contain malicious instructions designed to manipulate the agent's drafting process.
- Ingestion points: Content is fetched from public platforms including Reddit, Hacker News, and Bluesky via their respective public APIs and search endpoints.
- Boundary markers: The instructions do not define delimiters or specific "ignore instructions" directives to isolate the retrieved social media content from the agent's core instruction set.
- Capability inventory: The agent possesses the capability to execute shell commands (
curl) for data retrieval and is instructed to generate summaries and draft social media comments based on the retrieved data. - Sanitization: There is no evidence of content sanitization, filtering, or validation logic applied to the external data prior to its analysis and processing by the agent.
Audit Metadata