searching-group-leader_skill

This skill orchestrates sub-tasks to locate group leader information by coordinating General_Search_Agent and complementary agents for efficient search.
  • Python

2.5k

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill searching-group-leader

  • _meta.json298 B
  • SKILL.md1.4 KB

Overview

This skill organizes and leads a searching group by decomposing a search assignment into independent, complementary sub-tasks and assigning them to the most suitable agents. It always uses General_Search_Agent as the primary worker and can recommend additional specialized agents for specific subtasks. The goal is efficient parallel searches with clear responsibilities and minimal coordination overhead.

How this skill works

Given a search objective, the leader splits the work into up to eight independent sub-tasks that together cover the full scope. Each sub-task is described in plain language and is assigned to General_Search_Agent; optionally, the leader recommends other agents for specialized roles (e.g., web, legal, code search) according to query type. The leader performs a single parallel assignment pass and enforces constraints, such as avoiding Academic_Search for date-specific queries.

When to use it

  • Complex searches that benefit from parallelization and role separation
  • Tasks requiring coverage across different domains (news, code, documentation, forums)
  • Situations where a single coordinated plan and one-shot task assignment are needed
  • Searches with a broad scope that must be split into independent parts
  • When you want clear, agent-level responsibility and deliverables

Best practices

  • Limit sub-tasks to as few as necessary, up to eight, to reduce coordination cost
  • Always assign each sub-task to General_Search_Agent and clearly state its deliverable format
  • Add specialized agents only when their domain expertise improves results (e.g., Legal_Search for statutes)
  • Avoid calling Academic_Search for date-sensitive searches per policy
  • Provide explicit scope, sources to include/exclude, and time bounds for each sub-task

Example use cases

  • Gathering recent product mentions across news, forums, and social media: split into news, forums, social, and sentiment analysis sub-tasks
  • Reconciling documentation across versions: assign versioned doc retrieval, change extraction, and summary tasks
  • Competitive intelligence: web intel, job postings, patent search, and pricing signals as separate sub-tasks
  • Bug triage research: reproduce steps, search issue trackers, collect related stackoverflow threads, and summarize fixes

FAQ

Yes. General_Search_Agent must be used for every sub-task; additional agents are optional recommendations only.

How many sub-tasks should I create?

Use as few as necessary for clear separation of concerns, up to a hard limit of eight. Aim for 3–6 for most problems.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational