end-to-end-leadership-change-signal
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes local Python scripts within the project structure (e.g.,
skills/twitter-scraper/scripts/search_twitter.py) to fetch data from various platforms. This execution is part of the core discovery functionality. - [DATA_EXPOSURE]: The skill requests and stores Apollo API keys in a local JSON configuration file (
clients/<client-name>/config/end-to-end-leadership-change-signal.json). This is a standard and safe method for the agent to manage necessary service credentials. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from web searches and social media to generate outreach drafts, creating a potential surface for indirect injection.
- Ingestion points: Web search results, Tweets, Reddit posts, and Hacker News threads (referenced in Step 1).
- Boundary markers: No explicit delimiters or boundary instructions are mentioned for the external data processing steps.
- Capability inventory: Subprocess calls (via local scripts), file-writing (CSV/Markdown), network operations (Apollo API), and chaining to outreach tools.
- Sanitization: The instructions do not define specific sanitization or validation routines for the scraped web content before it is used by the LLM for email drafting.
Audit Metadata