Search Analyst Pair

Turn any research request into a structured, reviewable brief — fact collection, risk analysis, and recommendation in three deterministic steps.

Security Scan

The skill's requirements and instructions are internally consistent with a multi-agent `/hunt` orchestration workflow, but it asks for a high‑privilege OpenClaw gateway token and has a few ambiguous notes (fallback runner, local path) you should review before installing.

Purpose & Capability

Name/description (deterministic multi-agent research workflow) matches the declared requirements: OPENCLAW_BASE_URL and OPENCLAW_TOKEN are needed to call the OpenClaw gateway and spawn agents; curl/jq/python3 are reasonable for making and parsing HTTP calls and running an optional fallback runner.

Instruction Scope

SKILL.md confines behavior to the described Search->Analyst->Main chain and requires agent-to-agent permissions. It does not instruct reading arbitrary system files or exfiltrating data. However, it references support for a 'FastAPI fallback runner' while no code is included, and README contains a developer-local filesystem path—both are ambiguous and worth asking the author to clarify (where would fallback code run, what would it do?).

Install Mechanism

Instruction-only skill with no install spec or remote downloads; this is lowest-risk from an installation perspective because nothing is written to disk by the package itself.

Credentials

The single required credential (OPENCLAW_TOKEN) is appropriate for orchestrating agents, but that token can be high‑privilege (control over agents and actions). Confirm the token's scope/least-privilege policy before granting it to the skill.

Persistence & Privilege

always:false and normal autonomous invocation; the skill does not request permanent/platform-level presence and does not modify other skills. No elevated persistence is requested.

Assessment

This skill appears to do what it says: coordinate a fixed Search->Analyst->Main workflow via your OpenClaw gateway. Before installing: (1) ensure OPENCLAW_TOKEN has the minimum scope needed (prefer a token limited to spawning/communicating with the specific subagents rather than a full‑admin token); (2) verify your OpenClaw agent-to-agent allowlists and tools.agentToAgent settings; (3) ask the publisher what the 'FastAPI fallback runner' entails and where it would run (there's no code in the package); (4) treat the token as sensitive—do not paste it into third‑party systems—and test in a staging environment first. If you need higher assurance, request the fallback runner code and a description of the exact API calls the skill will make so you can audit them.

This skill packages a deterministic /hunt workflow for OpenClaw.

When a user message starts with /hunt, the workflow follows a strict 3-hop chain:

1. Search (DeepSeek): gather facts and sources only.
2. Analyst (Gemini): analyze Search output and identify key points and risks.
3. Main (Kimi): synthesize Search + Analyst into final guidance.

Why this skill

• Prevents ad-hoc routing for critical tasks.
• Separates fact collection, analysis, and decision output.
• Supports both native OpenClaw orchestration and a FastAPI fallback runner.

Best-fit scenarios

• Time-sensitive research tasks that require reliable structure.
• Decision support where source traceability matters.
• Team workflows that need stable, reviewable output sections.

Requirements

• OpenClaw gateway running and reachable.
• Agent IDs available: main, search, analyst.
• tools.agentToAgent.enabled=true in OpenClaw config.
• subagents.allowAgents configured:
  • main allows search, analyst
  • search allows analyst

Behavior contract

• Trigger prefix: /hunt
• Fixed order: Search -> Analyst -> Main
• Fallback policy: if agent-to-agent spawn fails, the workflow must explicitly mark fallback output.

Usage examples

/hunt Review today's agent framework updates and give a practical migration plan.

/hunt Collect top legal AI workflow changes this week and assess implementation risk.

Output shape (recommended)

• Search findings
• Analyst assessment
• Main conclusion