Help us improve
Share bugs, ideas, or general feedback.
From searchnet
Deep investigation of a single topic using iterative depth-first search. Best for "how does X work?" and "explain Y thoroughly" questions.
npx claudepluginhub pablolion/searchnet --plugin searchnetHow this skill is triggered — by the user, by Claude, or both
Slash command
/searchnet:deep-searchThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
You are the deep search orchestrator. This pipeline answers mechanism-oriented
Decomposes research questions into a DAG of sub-questions, executes parallel subagent searches, iterates on gaps, and synthesizes a final report. Useful for thorough, structured research on complex topics.
Enables recursive research: takes a query, searches, reads sources, extracts info, and deepens understanding through iterative exploration.
Conducts deep web research with parallel agents, multi-wave exploration for gaps, and structured synthesis. Activates for investigating topics, comparing options, best practices, or comprehensive web info.
Share bugs, ideas, or general feedback.
You are the deep search orchestrator. This pipeline answers mechanism-oriented questions: "how does X work?", "explain Y in detail", "what's the internals of Z?"
Before starting the search, present the user with:
Ask: "This will launch ~10 agents. Proceed?"
Only continue to Step 1 after user confirms.
workers/ and leaders/ subdirectoriessearch-plan.mdLaunch 4-6 searchnet:opus-searcher agents (one per subtopic), in parallel
(background).
Each worker:
workers/deep-{subtopic}.mdAfter all workers complete, launch 2 searchnet:opus-searcher agents in
parallel (background).
Each leader:
leaders/leader-{A..B}.mdRead both leader reports. Produce a coherent narrative explaining the full
mechanism. Write to final-report.md.
Launch the searchnet:verifier agent on final-report.md.