startup-launch-research

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests user-generated data from Product Hunt and Kickstarter (e.g., reviews, comments, and updates). This untrusted content could potentially contain hidden instructions designed to manipulate the AI agent's behavior.
  • Ingestion points: Data returned by the Crawlora API via scripts/crawlora.sh, specifically from the /producthunt/product/{id}/reviews and /kickstarter/comments endpoints.
  • Boundary markers: The skill does not currently employ specific markers or delimiters to isolate untrusted content from the agent's instructions.
  • Capability inventory: The skill possesses network access to the vendor's API via GET requests.
  • Sanitization: While the helper script sanitizes input parameters (paths and query strings), it does not sanitize the actual content of the retrieved reviews or comments.
  • [COMMAND_EXECUTION]: The skill utilizes a bash script (scripts/crawlora.sh) to perform API requests. The script is highly defensive, incorporating character-set restrictions for API keys, strict path whitelisting via regular expressions, and the use of temporary configuration files to prevent API keys from appearing in the system's process list.
  • [EXTERNAL_DOWNLOADS]: The skill makes network requests to https://api.crawlora.net/api/v1 to fetch startup research data. This domain belongs to the skill's author (crawlora-org), and the communication is restricted to secure GET requests.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 02:44 AM
Security Audit — agent-trust-hub — startup-launch-research