Skills
Navigate
skills/fivetaku/insane-research/insane-research-main/Gen Agent Trust Hub

insane-research-main

Pass

Audited by Gen Agent Trust Hub on Oct 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The SKILL.md file executes a setup script located at ${CLAUDE_PLUGIN_ROOT}/setup/setup.sh during initialization. Additionally, it deploys sub-agents using the Task tool with mode: \"bypassPermissions\", allowing them to execute tools and commands without individual user confirmation prompts.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external web sources. Ingestion points: Data enters via WebSearch, WebFetch, and curl commands in tool_strategy.md. Boundary markers: Rule R8 instructs agents to treat content as 'UNTRUSTED WEB CONTENT' and ignore instructions within it. Capability inventory: Uses Task spawning, Python scripts for verification, and has file system and network access. Sanitization: Relies on instructional constraints to prevent command execution from data.
  • [EXTERNAL_DOWNLOADS]: The tool_strategy.md file specifies data fetching from platforms including Jina Reader (r.jina.ai), X (syndication.twitter.com), and Reddit (.rss feeds) for information gathering and research purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 13, 2026, 10:31 AM