Doing A Simple Two Stage Fanout

Fan out a large corpus across Worker, Critic, and Summarizer subagents with computed layout

Tested · Works

Test report

Verdict
Tested · Works
Score
8.8/10
Tested
Jul 31, 2026
Environment
Claude Code 2.x (agent harness)
Upstream re-checked
Aug 10, 2026 · 13e70cd

Fetched SKILL.md plus both referenced files (compute_layout.py and diagram-templates.md, each HTTP 200); the script scanned clean (no curl|sh, base64, network, or eval). Ran compute_layout.py on the SKILL's own documented example (--corpus-chars 800000 --segments-per 3 --reviews-per 2) and got exactly the numbers the skill claims: 6 segments, 2 workers, 4 critics (C01-C04) matching its assignment table. OUTPUT test on a 500k-char prose corpus: baseline naive reasoning (125k tokens fits in a 200k window) gives 1 agent / no fan-out, whereas the skill's script reserves 35% and divides by SEGMENTS_PER to yield 4 segments / 2 workers / 3 critics / 6 agents - a concrete, defensible divergence I verified by running it. I did not execute the full multi-subagent dispatch end-to-end (no live Worker/Critic/Summarizer run), only the deterministic layout-computation slice.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

  • Installs cleanly 5/5
  • Triggers reliably 5/5
  • Output vs. baseline 7/10
  • Docs & honesty 5/5

What Doing A Simple Two Stage Fanout does

Orchestrates a two-stage fan-out for analyzing a corpus larger than one agent's context: parallel Worker subagents each analyze a slice, Critic subagents review the worker reports for gaps and contradictions, and a single Summarizer synthesizes a final report. Triggers when the user asks to analyze, review, or summarize a large body of text, code, or data that exceeds a single agent's context, and includes a bundled Python script that computes segment/worker/critic counts plus task-graph and failure-recovery guidance.

How to install Doing A Simple Two Stage Fanout

git clone --depth 1 https://github.com/ed3dai/ed3d-plugins.git /tmp/doing-a-simple-two-stage-fanout-src
mkdir -p ~/.claude/skills
cp -R /tmp/doing-a-simple-two-stage-fanout-src/plugins/ed3d-basic-agents/skills/doing-a-simple-two-stage-fanout ~/.claude/skills/doing-a-simple-two-stage-fanout
# Bundled compute_layout.py needs only python3 (stdlib argparse/math/os) - no pip deps.
# Runtime needs: Task tool (subagent dispatch), TaskCreate/TaskUpdate, AskUserQuestion, plan mode.
# Plugin-marketplace alternative: add the ed3dai/ed3d-plugins marketplace, install the ed3d-basic-agents plugin.

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Doing A Simple Two Stage Fanout

  • /doing-a-simple-two-stage-fanout Fan out a large corpus across Worker, Critic, and Summarizer subagents with computed layout

It also activates on plain-language prompts like these:

  • Analyze this entire codebase across many subagents
  • Fan out this large dataset to worker agents
  • Summarize this huge document set in parallel

Frequently asked questions

Is the Doing A Simple Two Stage Fanout skill free?
Yes. The skill itself is free from ed3dai/ed3d-plugins. SkillProof publishes the install command and an independent test verdict at no cost.
Does Doing A Simple Two Stage Fanout work with Claude Code?
We tested it with Claude Code 2.x (agent harness) on Jul 31, 2026. Verdict: Tested · Works. Fetched SKILL.md plus both referenced files (compute_layout.py and diagram-templates.md, each HTTP 200); the script scanned clean (no curl|sh, base64, network, or eval). Ran compute_layout.py on the SKILL's own documented example (--corpus-chars 800000 --segments-per 3 --reviews-per 2) and got exactly the numbers the skill claims: 6 segments, 2 workers, 4 critics (C01-C04) matching its assignment table. OUTPUT test on a 500k-char prose corpus: baseline naive reasoning (125k tokens fits in a 200k window) gives 1 agent / no fan-out, whereas the skill's script reserves 35% and divides by SEGMENTS_PER to yield 4 segments / 2 workers / 3 critics / 6 agents - a concrete, defensible divergence I verified by running it. I did not execute the full multi-subagent dispatch end-to-end (no live Worker/Critic/Summarizer run), only the deterministic layout-computation slice.
What is the Doing A Simple Two Stage Fanout SkillProof Score?
8.8/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 7/10, docs & honesty 5/5.
How do I install Doing A Simple Two Stage Fanout?
Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Doing A Simple Two Stage Fanout with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.