show-hn-writer
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill includes a Python script in
SKILL.md(Step 8) designed to be executed by the agent. This script performs network requests and local file writing (hn_log.csv). While intended for scraping Hacker News data, it represents a dynamic execution path using the agent's interpreter. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted external data from project README files and user-provided context to draft posts.
- Ingestion points: Reads project context from local files (README.md) and user responses in Step 1.
- Boundary markers: No delimiters or instructions are used to distinguish untrusted data from the skill's own logic.
- Capability inventory: The skill uses Python execution (via
requestsandThreadPoolExecutor) and interacts with the Gemini API for quality reviews. - Sanitization: There is no evidence of validation or escaping for the data interpolated into the drafting process or the Python scraper's keyword field.
- [EXTERNAL_DOWNLOADS]: The included Python script fetches data from the Hacker News API at
https://hacker-news.firebaseio.com/v0. This is a well-known service used for legitimate data retrieval.
Audit Metadata