Scout — Research Agent
Scout — Research Agent
What it is: The research agent — given a topic, investigates across sources, synthesizes findings, and returns a sourced intelligence report.
Who it's for: Founders, analysts, and developers who need fast, reliable research without spending hours manually gathering information.
What it does:
- Searches across web, documentation, and specified sources
- Synthesizes findings into structured, sourced reports
- Tracks information provenance — every claim linked to a source
- Identifies gaps and contradictions in gathered data
- Delivers structured output ready for downstream agents to act on
What it won't do:
- Does not fabricate sources or cite unverified information
- Does not take direct actions based on research — report outputs only
- Does not access authenticated resources without explicit permission
Needs:
- Research query or topic
- Optional: source list or domain restrictions
- Web access governed by web_access.json policy
Produces:
- Sourced research report
- Key findings summary
- Source credibility assessment
- Gap and contradiction analysis
Install in 60 seconds: Extract ZIP, run installer. Full guide: https://hivedeck.ai/pages/how-to-install
Works with: OpenClaw, Claude Code, Cursor, AutoGPT, CrewAI, LangGraph
Scout operates under Tier 1 Governance — every action bounded, logged, and auditable
Stop Googling. Let Scout research for you.
Scout is a fully autonomous research agent for OpenClaw. Deep competitive analysis, technology evaluations, market scans, Porter's Five Forces — with proper citations — dropped straight into your Telegram or WhatsApp.
One-command install. Runs on a schedule. Never goes rogue — operates under your Governance agents (Sentinel/Auditor) so every action is bounded and auditable.
What you actually get (12 files and folders, ready in under 60 seconds)
- install.sh / uninstall.sh / verify.sh — one-command setup and teardown
- skills/ — full research-specific tool definitions
- SOUL.md — Scout's complete personality and system prompt
- TOOLS.md — every tool Scout uses, pre-configured
- HEARTBEAT.md + cron/ — scheduled daily and weekly research runs (it stays alive automatically)
- config/ / scripts/ / docs/ — everything pre-configured, nothing to guess
Install in 4 steps
- Download the ZIP
cd ~/.openclaw/agents && unzip scout.zip./install.sh- Tell your Chief of Staff: "Route all research to Scout"
Works with Claude 3.5/4 or GPT-5. Fully compatible with OpenClaw's governance layer — Sentinel can veto, Auditor can review, everything is logged.
What Scout does on a schedule
- Competitive landscape analysis with citations
- Technology evaluation reports
- Market assessment and sizing
- Porter's Five Forces breakdowns
- Weekly research digests delivered to your Telegram/WhatsApp
- Ad hoc deep dives on demand
Built for OpenClaw. Safe by design.
After the ClawHavoc community skills incident, running untested agents from random sources is a real risk. Scout ships with the same governance architecture we use in our own 18-agent production system. It cannot operate outside the boundaries you set. Every action is logged. You stay in control.
Part of the Execution Stack ($149 — save $66 vs. buying separately).
New to HiveDeck? Step-by-step install guide →
Related agents: Ranker — SEO Command Agent | Prism — Brand Strategist | Chief of Staff — Routing Authority
Install guide: hivedeck.ai/pages/how-to-install