Not in the top 1,000 by GitHub stars

Ali-expandings/wise-men-flash

The fast tier of the wise-men council as its own skill: three Claude subagents down three reasoning paths, one checked decision memo in a third of the full council's time. Ships its own blind eval.

History

Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.

View weekly changes

SKILL.md

Week of Nov 2, 2026 · +0 / −0 lines

Before Not checked yet → After Local scripts

aaabd74f8f6d → b8db7ddada08

Diff text is unavailable under the recorded licenses. Line counts and the GitHub link remain available.

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Not checked yet

First observation → aaabd74f8f6d

No earlier version was observed for comparison.

Compare on GitHub

View on GitHubAdd to my skills

Verified commit b8db7ddada08275fb16c25e526a84823e62a85fa

git clone --filter=blob:none --no-checkout https://github.com/Ali-expandings/wise-men-flash.git .hotskills/ali-expandings-wise-men-flash && git -C .hotskills/ali-expandings-wise-men-flash checkout --detach b8db7ddada08275fb16c25e526a84823e62a85fa # Copy each selected skill directory from .hotskills/ali-expandings-wise-men-flash into .claude/skills/<skill-name>/
git clone --filter=blob:none --no-checkout https://github.com/Ali-expandings/wise-men-flash.git .hotskills/ali-expandings-wise-men-flash && git -C .hotskills/ali-expandings-wise-men-flash checkout --detach b8db7ddada08275fb16c25e526a84823e62a85fa # Copy each selected skill directory from .hotskills/ali-expandings-wise-men-flash into .agents/skills/<skill-name>/
10GitHub stars
1star in 7 days
1skill in this repository
6commits in 30 days

Heat

4.5
  • Star growth, last 7 daysNot available
  • Star growth, last 24 hoursNot available
  • Rank movement, last 7 daysNot available
  • Commits, last 30 daysNot available
  • Accounts asking about it through MCP, last 7 daysNot available
  • Hot votes this weekNot available

Heat never changes the order by stars, and sponsorship never touches it. How heat works

Stars over time

Oct 28, 2026Nov 6, 202610

30-day change not yet available: not enough history.

Individual skills in this repository

How it works

wise-men-flash

When your agent uses it

Use when the user asks for a "flash council", "quick council", "fast panel", "wise men flash", "quick second opinion from a panel", or invokes /wise-men-flash — and when they want a council-quality answer to a hard question fast (four subagent calls and about a third of the full council's time, a median 16.7 minutes in its own pre-registered round) rather than the full wise-men protocol. A three-member council of a practitioner anchor named for the job the question belongs to, one domain member, and a Devil's Advocate — each reasoning down a different path, spawned in parallel. No peer-review round, no debate; the main thread synthesizes one decision memo, and a single fresh checker verifies it before the user sees it. Members are offline (Read/Grep/Glob only). Its answer is a decision memo — Recommendation / Why / What to do / Risks / Strongest counter-position / Confidence — with no council mechanics in it. Sibling of the full `wise-men` skill, with the same answer format and one fixed fast tier instead of five. Pure Claude — no external APIs.

What it loads into context

Frontmatter

252 tokens

Always available to the agent

SKILL.md

5,523 tokens

When the skill is activated

Reference documents

915,046 tokens

Loaded on request

Token counts are estimates. Scripts run separately; they are not counted as reference context.

The playbook

  1. Orchestrator runbook (execute this; the reference sections below explain it)
  2. Pre-flight (the four checks in full)
  3. Stage 0 — roster selection
  4. Stage 0.5 — the shared brief
  5. Stage 1 — the member prompt
  6. Stage 4 — the decision memo (the only output)
  7. Stage 4.5 — the checks
  8. Model routing
  9. Honest limits
  10. Credits

Scripts executed separately

  • eval-data/head-to-head/flash4.py
  • eval-data/head-to-head/h2h5.py
  • scripts/check.sh
  • scripts/make_charts.py

Files

  • .claude-plugin/marketplace.jsonasset668 bytes
  • .claude-plugin/plugin.jsonasset677 bytes
  • .github/workflows/check.ymlasset311 bytes
  • .gitignoreasset114 bytes
  • AGENTS.mddoc794 bytes
  • CHANGELOG.mddoc6,289 bytes
  • CONTRIBUTING.mddoc1,293 bytes
  • DETAILS.mddoc20,839 bytes
  • LICENSEasset1,060 bytes
  • README.mddoc5,494 bytes
  • SECURITY.mddoc644 bytes
  • SKILL.mdcore24,364 bytes
  • SKILL.mdcore24,364 bytes
  • SKILL.mdcore24,364 bytes
  • agents/wise-member.mddoc1,199 bytes
  • assets/banner.svgasset30,958 bytes
  • assets/round4-axes-dark.svgasset11,430 bytes
  • assets/round4-axes.svgasset11,430 bytes
  • assets/round4-dark.svgasset6,151 bytes
  • assets/round4-questions-dark.svgasset7,564 bytes
  • assets/round4-questions.svgasset7,564 bytes
  • assets/round4-speed-dark.svgasset7,665 bytes
  • assets/round4-speed.svgasset7,665 bytes
  • assets/round4.svgasset6,151 bytes
  • assets/round5-axes-dark.svgasset18,512 bytes
  • assets/round5-axes.svgasset18,512 bytes
  • assets/round5-dark.svgasset7,787 bytes
  • assets/round5-questions-dark.svgasset13,144 bytes
  • assets/round5-questions.svgasset13,144 bytes
  • assets/round5-speed-dark.svgasset10,865 bytes
  • assets/round5-speed.svgasset10,865 bytes
  • assets/round5.svgasset7,787 bytes
  • eval-data/PREREG-1.mddoc5,373 bytes
  • eval-data/README.mddoc4,116 bytes
  • eval-data/consistency-check/PLAN.mddoc2,574 bytes
  • eval-data/consistency-check/RESULTS.mddoc2,361 bytes
  • eval-data/consistency-check/answers/R502-new.mddoc10,122 bytes
  • eval-data/consistency-check/answers/R504-new.mddoc9,347 bytes
  • eval-data/consistency-check/answers/R506-new.mddoc8,707 bytes
  • eval-data/consistency-check/answers/R508-new.mddoc9,187 bytes
  • eval-data/consistency-check/blinded/R502-j1.mddoc21,143 bytes
  • eval-data/consistency-check/blinded/R502-j2.mddoc21,143 bytes
  • eval-data/consistency-check/blinded/R504-j1.mddoc20,088 bytes
  • eval-data/consistency-check/blinded/R504-j2.mddoc20,088 bytes
  • eval-data/consistency-check/blinded/R506-j1.mddoc20,399 bytes
  • eval-data/consistency-check/blinded/R506-j2.mddoc20,399 bytes
  • eval-data/consistency-check/blinded/R508-j1.mddoc20,284 bytes
  • eval-data/consistency-check/blinded/R508-j2.mddoc20,284 bytes
  • eval-data/consistency-check/classification.mddoc2,436 bytes
  • eval-data/consistency-check/judgments/R502-j1.mddoc4,305 bytes
  • eval-data/consistency-check/judgments/R502-j2.mddoc4,035 bytes
  • eval-data/consistency-check/judgments/R504-j1.mddoc4,736 bytes
  • eval-data/consistency-check/judgments/R504-j2.mddoc4,759 bytes
  • eval-data/consistency-check/judgments/R506-j1.mddoc3,853 bytes
  • eval-data/consistency-check/judgments/R506-j2.mddoc3,822 bytes
  • eval-data/consistency-check/judgments/R508-j1.mddoc4,939 bytes
  • eval-data/consistency-check/judgments/R508-j2.mddoc4,179 bytes
  • eval-data/consistency-check/order.txtdoc72 bytes
  • eval-data/head-to-head/PREREG-4.mddoc6,502 bytes
  • eval-data/head-to-head/PREREG-5.mddoc8,389 bytes
  • eval-data/head-to-head/RESULTS-V5.mddoc11,753 bytes
  • eval-data/head-to-head/RESULTS.mddoc8,066 bytes
  • eval-data/head-to-head/arm-prompts-r5.mddoc11,175 bytes
  • eval-data/head-to-head/blinded-v4/CAL-Q23.mddoc32,700 bytes
  • eval-data/head-to-head/blinded-v4/R401-j1.mddoc44,042 bytes
  • eval-data/head-to-head/blinded-v4/R401-j2.mddoc44,042 bytes
  • eval-data/head-to-head/blinded-v4/R401-j3.mddoc44,042 bytes
  • eval-data/head-to-head/blinded-v4/R402-j1.mddoc40,418 bytes
  • eval-data/head-to-head/blinded-v4/R402-j2.mddoc40,418 bytes
  • eval-data/head-to-head/blinded-v4/R402-j3.mddoc40,418 bytes
  • eval-data/head-to-head/blinded-v4/R403-j1.mddoc37,356 bytes
  • eval-data/head-to-head/blinded-v4/R403-j2.mddoc37,356 bytes
  • eval-data/head-to-head/blinded-v4/R403-j3.mddoc37,356 bytes
  • eval-data/head-to-head/blinded-v4/R404-j1.mddoc34,766 bytes
  • eval-data/head-to-head/blinded-v4/R404-j2.mddoc34,766 bytes
  • eval-data/head-to-head/blinded-v4/R404-j3.mddoc34,766 bytes
  • eval-data/head-to-head/blinded-v4/R405-j1.mddoc42,071 bytes
  • eval-data/head-to-head/blinded-v4/R405-j2.mddoc42,071 bytes
  • eval-data/head-to-head/blinded-v4/R405-j3.mddoc42,071 bytes
  • eval-data/head-to-head/blinded-v5/R501-j1.mddoc91,458 bytes
  • eval-data/head-to-head/blinded-v5/R501-j2.mddoc91,458 bytes
  • eval-data/head-to-head/blinded-v5/R501-j3.mddoc91,458 bytes
  • eval-data/head-to-head/blinded-v5/R502-j1.mddoc79,988 bytes
  • eval-data/head-to-head/blinded-v5/R502-j2.mddoc79,988 bytes
  • eval-data/head-to-head/blinded-v5/R502-j3.mddoc79,988 bytes
  • eval-data/head-to-head/blinded-v5/R503-j1.mddoc83,757 bytes
  • eval-data/head-to-head/blinded-v5/R503-j2.mddoc83,757 bytes
  • eval-data/head-to-head/blinded-v5/R503-j3.mddoc83,757 bytes
  • eval-data/head-to-head/blinded-v5/R504-j1.mddoc72,396 bytes
  • eval-data/head-to-head/blinded-v5/R504-j2.mddoc72,396 bytes
  • eval-data/head-to-head/blinded-v5/R504-j3.mddoc72,396 bytes
  • eval-data/head-to-head/blinded-v5/R505-j1.mddoc82,252 bytes
  • eval-data/head-to-head/blinded-v5/R505-j2.mddoc82,252 bytes
  • eval-data/head-to-head/blinded-v5/R505-j3.mddoc82,252 bytes
  • eval-data/head-to-head/blinded-v5/R506-j1.mddoc71,295 bytes
  • eval-data/head-to-head/blinded-v5/R506-j2.mddoc71,295 bytes
  • eval-data/head-to-head/blinded-v5/R506-j3.mddoc71,295 bytes
  • eval-data/head-to-head/blinded-v5/R507-j1.mddoc88,381 bytes
  • eval-data/head-to-head/blinded-v5/R507-j2.mddoc88,381 bytes
  • eval-data/head-to-head/blinded-v5/R507-j3.mddoc88,381 bytes
  • eval-data/head-to-head/blinded-v5/R508-j1.mddoc92,421 bytes
  • eval-data/head-to-head/blinded-v5/R508-j2.mddoc92,421 bytes
  • eval-data/head-to-head/blinded-v5/R508-j3.mddoc92,421 bytes
  • eval-data/head-to-head/blinding4.yamlasset2,999 bytes
  • eval-data/head-to-head/blinding5.yamlasset4,967 bytes
  • eval-data/head-to-head/flash4.pyruns19,619 bytes
  • eval-data/head-to-head/h2h5.pyruns24,024 bytes
  • eval-data/head-to-head/judge-prompt-5.txtdoc1,737 bytes
  • eval-data/head-to-head/judge-prompt-r5.txtdoc1,749 bytes
  • eval-data/head-to-head/judgments-v4/CAL-Q23.mddoc8,504 bytes
  • eval-data/head-to-head/judgments-v4/R401-j1.mddoc10,458 bytes
  • eval-data/head-to-head/judgments-v4/R401-j2.mddoc9,574 bytes
  • eval-data/head-to-head/judgments-v4/R401-j3.mddoc8,368 bytes
  • eval-data/head-to-head/judgments-v4/R402-j1.mddoc9,376 bytes
  • eval-data/head-to-head/judgments-v4/R402-j2.mddoc10,103 bytes
  • eval-data/head-to-head/judgments-v4/R402-j3.mddoc10,172 bytes
  • eval-data/head-to-head/judgments-v4/R403-j1.mddoc9,642 bytes
  • eval-data/head-to-head/judgments-v4/R403-j2.mddoc9,170 bytes
  • eval-data/head-to-head/judgments-v4/R403-j3.mddoc8,461 bytes
  • eval-data/head-to-head/judgments-v4/R404-j1.mddoc9,896 bytes
  • eval-data/head-to-head/judgments-v4/R404-j2.mddoc9,907 bytes
  • eval-data/head-to-head/judgments-v4/R404-j3.mddoc9,839 bytes
  • eval-data/head-to-head/judgments-v4/R405-j1.mddoc8,970 bytes
  • eval-data/head-to-head/judgments-v4/R405-j2.mddoc8,962 bytes
  • eval-data/head-to-head/judgments-v4/R405-j3.mddoc9,673 bytes
  • eval-data/head-to-head/judgments-v5/R501-j1.mddoc11,737 bytes
  • eval-data/head-to-head/judgments-v5/R501-j2.mddoc12,631 bytes
  • eval-data/head-to-head/judgments-v5/R501-j3.mddoc15,871 bytes
  • eval-data/head-to-head/judgments-v5/R502-j1.mddoc15,110 bytes
  • eval-data/head-to-head/judgments-v5/R502-j2.mddoc13,352 bytes
  • eval-data/head-to-head/judgments-v5/R502-j3.mddoc14,832 bytes
  • eval-data/head-to-head/judgments-v5/R503-j1.mddoc13,316 bytes
  • eval-data/head-to-head/judgments-v5/R503-j2.mddoc14,828 bytes
  • eval-data/head-to-head/judgments-v5/R503-j3.mddoc14,023 bytes
  • eval-data/head-to-head/judgments-v5/R504-j1.mddoc13,313 bytes
  • eval-data/head-to-head/judgments-v5/R504-j2.mddoc14,288 bytes
  • eval-data/head-to-head/judgments-v5/R504-j3.mddoc12,784 bytes
  • eval-data/head-to-head/judgments-v5/R505-j1.mddoc13,561 bytes
  • eval-data/head-to-head/judgments-v5/R505-j2.mddoc12,168 bytes
  • eval-data/head-to-head/judgments-v5/R505-j3.mddoc11,657 bytes
  • eval-data/head-to-head/judgments-v5/R506-j1.mddoc12,995 bytes
  • eval-data/head-to-head/judgments-v5/R506-j2.mddoc13,616 bytes
  • eval-data/head-to-head/judgments-v5/R506-j3.mddoc12,992 bytes
  • eval-data/head-to-head/judgments-v5/R507-j1.mddoc14,122 bytes
  • eval-data/head-to-head/judgments-v5/R507-j2.mddoc14,804 bytes
  • eval-data/head-to-head/judgments-v5/R507-j3.mddoc11,848 bytes
  • eval-data/head-to-head/judgments-v5/R508-j1.mddoc14,679 bytes
  • eval-data/head-to-head/judgments-v5/R508-j2.mddoc11,033 bytes
  • eval-data/head-to-head/judgments-v5/R508-j3.mddoc15,441 bytes
  • eval-data/head-to-head/parsed-v4/R401.yamlasset1,934 bytes
  • eval-data/head-to-head/parsed-v4/R402.yamlasset1,934 bytes
  • eval-data/head-to-head/parsed-v4/R403.yamlasset1,934 bytes
  • eval-data/head-to-head/parsed-v4/R404.yamlasset1,934 bytes
  • eval-data/head-to-head/parsed-v4/R405.yamlasset1,934 bytes
  • eval-data/head-to-head/parsed-v5/R501.yamlasset3,428 bytes
  • eval-data/head-to-head/parsed-v5/R502.yamlasset3,428 bytes
  • eval-data/head-to-head/parsed-v5/R503.yamlasset3,428 bytes
  • eval-data/head-to-head/parsed-v5/R504.yamlasset3,428 bytes
  • eval-data/head-to-head/parsed-v5/R505.yamlasset3,428 bytes
  • eval-data/head-to-head/parsed-v5/R506.yamlasset3,428 bytes
  • eval-data/head-to-head/parsed-v5/R507.yamlasset3,428 bytes
  • eval-data/head-to-head/parsed-v5/R508.yamlasset3,428 bytes
  • eval-data/head-to-head/question-author-prompt-5.txtdoc3,509 bytes
  • eval-data/head-to-head/question-author-prompt.txtdoc3,022 bytes
  • eval-data/head-to-head/questions-r4.yamlasset3,329 bytes
  • eval-data/head-to-head/questions-r5.yamlasset4,703 bytes
  • eval-data/head-to-head/raw/R401/direct.mddoc4,468 bytes
  • eval-data/head-to-head/raw/R401/llm-council.mddoc8,998 bytes
  • eval-data/head-to-head/raw/R401/warp-council.mddoc7,047 bytes
  • eval-data/head-to-head/raw/R401/wise-men-3.13-fast.mddoc8,937 bytes
  • eval-data/head-to-head/raw/R401/wise-men-3.13.mddoc13,901 bytes
  • eval-data/head-to-head/raw/R402/direct.mddoc3,393 bytes
  • eval-data/head-to-head/raw/R402/llm-council.mddoc9,662 bytes
  • eval-data/head-to-head/raw/R402/warp-council.mddoc5,807 bytes
  • eval-data/head-to-head/raw/R402/wise-men-3.13-fast.mddoc9,469 bytes
  • eval-data/head-to-head/raw/R402/wise-men-3.13.mddoc13,305 bytes
  • eval-data/head-to-head/raw/R403/direct.mddoc3,650 bytes
  • eval-data/head-to-head/raw/R403/llm-council.mddoc5,192 bytes
  • eval-data/head-to-head/raw/R403/warp-council.mddoc6,630 bytes
  • eval-data/head-to-head/raw/R403/wise-men-3.13-fast.mddoc7,906 bytes
  • eval-data/head-to-head/raw/R403/wise-men-3.13.mddoc13,388 bytes
  • eval-data/head-to-head/raw/R404/direct.mddoc3,038 bytes
  • eval-data/head-to-head/raw/R404/llm-council.mddoc7,655 bytes
  • eval-data/head-to-head/raw/R404/warp-council.mddoc5,627 bytes
  • eval-data/head-to-head/raw/R404/wise-men-3.13-fast.mddoc8,591 bytes
  • eval-data/head-to-head/raw/R404/wise-men-3.13.mddoc9,277 bytes
  • eval-data/head-to-head/raw/R405/direct.mddoc4,228 bytes
  • eval-data/head-to-head/raw/R405/llm-council.mddoc10,749 bytes
  • eval-data/head-to-head/raw/R405/warp-council.mddoc5,956 bytes
  • eval-data/head-to-head/raw/R405/wise-men-3.13-fast.mddoc9,017 bytes
  • eval-data/head-to-head/raw/R405/wise-men-3.13.mddoc11,532 bytes
  • eval-data/head-to-head/raw/R501/brainstorming.mddoc8,097 bytes
  • eval-data/head-to-head/raw/R501/direct.mddoc5,973 bytes
  • eval-data/head-to-head/raw/R501/ecc-council.mddoc5,161 bytes
  • eval-data/head-to-head/raw/R501/grilling.mddoc13,658 bytes
  • eval-data/head-to-head/raw/R501/lifeos-council.mddoc16,757 bytes
  • eval-data/head-to-head/raw/R501/llm-council.mddoc8,991 bytes
  • eval-data/head-to-head/raw/R501/warp-council.mddoc7,806 bytes
  • eval-data/head-to-head/raw/R501/wise-men-3.14.mddoc14,597 bytes
  • eval-data/head-to-head/raw/R501/wise-men-flash.mddoc10,670 bytes
  • eval-data/head-to-head/raw/R502/brainstorming.mddoc11,295 bytes
  • eval-data/head-to-head/raw/R502/direct.mddoc3,739 bytes
  • eval-data/head-to-head/raw/R502/ecc-council.mddoc3,844 bytes
  • eval-data/head-to-head/raw/R502/grilling.mddoc9,630 bytes
  • eval-data/head-to-head/raw/R502/lifeos-council.mddoc16,349 bytes
  • eval-data/head-to-head/raw/R502/llm-council.mddoc8,563 bytes
  • eval-data/head-to-head/raw/R502/warp-council.mddoc4,652 bytes
  • eval-data/head-to-head/raw/R502/wise-men-3.14.mddoc13,251 bytes
  • eval-data/head-to-head/raw/R502/wise-men-flash.mddoc9,276 bytes
  • eval-data/head-to-head/raw/R503/brainstorming.mddoc17,821 bytes
  • eval-data/head-to-head/raw/R503/direct.mddoc3,174 bytes
  • eval-data/head-to-head/raw/R503/ecc-council.mddoc3,429 bytes
  • eval-data/head-to-head/raw/R503/grilling.mddoc10,724 bytes
  • eval-data/head-to-head/raw/R503/lifeos-council.mddoc16,096 bytes
  • eval-data/head-to-head/raw/R503/llm-council.mddoc7,893 bytes
  • eval-data/head-to-head/raw/R503/warp-council.mddoc4,367 bytes
  • eval-data/head-to-head/raw/R503/wise-men-3.14.mddoc11,435 bytes
  • eval-data/head-to-head/raw/R503/wise-men-flash.mddoc10,385 bytes
  • eval-data/head-to-head/raw/R504/brainstorming.mddoc5,548 bytes
  • eval-data/head-to-head/raw/R504/direct.mddoc3,752 bytes
  • eval-data/head-to-head/raw/R504/ecc-council.mddoc4,744 bytes
  • eval-data/head-to-head/raw/R504/grilling.mddoc7,987 bytes
  • eval-data/head-to-head/raw/R504/lifeos-council.mddoc16,272 bytes
  • eval-data/head-to-head/raw/R504/llm-council.mddoc7,953 bytes
  • eval-data/head-to-head/raw/R504/warp-council.mddoc5,675 bytes
  • eval-data/head-to-head/raw/R504/wise-men-3.14.mddoc12,265 bytes
  • eval-data/head-to-head/raw/R504/wise-men-flash.mddoc9,057 bytes
  • eval-data/head-to-head/raw/R505/brainstorming.mddoc5,836 bytes
  • eval-data/head-to-head/raw/R505/direct.mddoc5,771 bytes
  • eval-data/head-to-head/raw/R505/ecc-council.mddoc4,368 bytes
  • eval-data/head-to-head/raw/R505/grilling.mddoc9,463 bytes
  • eval-data/head-to-head/raw/R505/lifeos-council.mddoc16,925 bytes
  • eval-data/head-to-head/raw/R505/llm-council.mddoc9,172 bytes
  • eval-data/head-to-head/raw/R505/warp-council.mddoc7,070 bytes
  • eval-data/head-to-head/raw/R505/wise-men-3.14.mddoc12,612 bytes
  • eval-data/head-to-head/raw/R505/wise-men-flash.mddoc13,306 bytes
  • eval-data/head-to-head/raw/R506/brainstorming.mddoc5,141 bytes
  • eval-data/head-to-head/raw/R506/direct.mddoc3,831 bytes
  • eval-data/head-to-head/raw/R506/ecc-council.mddoc3,974 bytes
  • eval-data/head-to-head/raw/R506/grilling.mddoc6,383 bytes
  • eval-data/head-to-head/raw/R506/lifeos-council.mddoc15,815 bytes
  • eval-data/head-to-head/raw/R506/llm-council.mddoc9,435 bytes
  • eval-data/head-to-head/raw/R506/warp-council.mddoc3,770 bytes
  • eval-data/head-to-head/raw/R506/wise-men-3.14.mddoc13,841 bytes
  • eval-data/head-to-head/raw/R506/wise-men-flash.mddoc9,902 bytes
  • eval-data/head-to-head/raw/R507/brainstorming.mddoc14,706 bytes
  • eval-data/head-to-head/raw/R507/direct.mddoc4,253 bytes
  • eval-data/head-to-head/raw/R507/ecc-council.mddoc3,459 bytes
  • eval-data/head-to-head/raw/R507/grilling.mddoc8,614 bytes
  • eval-data/head-to-head/raw/R507/lifeos-council.mddoc16,401 bytes
  • eval-data/head-to-head/raw/R507/llm-council.mddoc8,857 bytes
  • eval-data/head-to-head/raw/R507/warp-council.mddoc6,401 bytes
  • eval-data/head-to-head/raw/R507/wise-men-3.14.mddoc15,223 bytes
  • eval-data/head-to-head/raw/R507/wise-men-flash.mddoc11,697 bytes
  • eval-data/head-to-head/raw/R508/brainstorming.mddoc20,016 bytes
  • eval-data/head-to-head/raw/R508/direct.mddoc4,521 bytes
  • eval-data/head-to-head/raw/R508/ecc-council.mddoc3,495 bytes
  • eval-data/head-to-head/raw/R508/grilling.mddoc12,439 bytes
  • eval-data/head-to-head/raw/R508/lifeos-council.mddoc18,736 bytes
  • eval-data/head-to-head/raw/R508/llm-council.mddoc8,733 bytes
  • eval-data/head-to-head/raw/R508/warp-council.mddoc5,784 bytes
  • eval-data/head-to-head/raw/R508/wise-men-3.14.mddoc10,648 bytes
  • eval-data/head-to-head/raw/R508/wise-men-flash.mddoc9,343 bytes
  • examples/engineering-choice.mddoc10,404 bytes
  • examples/trust-a-number.mddoc9,958 bytes
  • requirements.txtdoc208 bytes
  • scripts/check.shruns6,729 bytes
  • scripts/make_charts.pyruns34,808 bytes
Read SKILL.md

# wise-men-flash

The fast tier of the wise-men council, extracted into its own skill. Three members, no peer-review round, one synthesis check. It answers in four subagent calls and about a third of the full council's time. Measured as installed in the pre-registered round-5 head-to-head (eight new questions, nine arms, three blind judges each): 23.4/25, clearly ahead of all seven rivals — six other skills and a plain answer — on every question, and a quarter point behind the full `wise-men` council (23.6) with an interval that includes zero, in a median 16.7 minutes and 3.95 USD a question against its 50.2 minutes and 17.72 USD (list price). Round 4 had measured the same profile as the wise-men fast arm: 23.3/25 in 10.8 minutes. Use it when the decision is real but does not justify the full protocol, or when subagent budget is tight.

**When to reach for the full `wise-men` instead**: an irreversible or high-blast-radius call, a compound question that needs splitting, or anything where you want peer-review scoring and a debate round. Flash is one speed. It does not adapt its depth to the stakes — that is exactly what `wise-men` (solo → quick → standard → deep → paranoid) is for. When in doubt on a genuinely high-stakes, hard-to-reverse decision, spend the full council.

## Orchestrator runbook (execute this; the reference sections below explain it)

You (main thread) are the orchestrator and the Chairman. These steps are MECHANICAL. Every deviation MUST be disclosed in the final output's footer — a silently shortcut council is the fake-council anti-pattern.

0. **Pre-flight** (4 checks). If the user already invoked the skill, do NOT ask "want a council?" — they asked for one. If the question is trivial, one-line, a settled fact, or a choice reversible in minutes, say so and answer directly; a council is overkill. If it is compound, split it and say which part you are answering. If it hinges on a current external fact (a price, a version, a tool's present state), gather that fact ONCE now and put it in the brief with its date — never let offline members answer time-sensitive questions from training data.
1. **Roster** (Stage 0): three members.
   - **Practitioner anchor** — named for the actual job the question belongs to (an SRE lead, an employment lawyer, a pricing lead, a launch DevRel lead). Owns correctness and completeness: the concrete first step and the must-check facts. **Strong model.**
   - **One domain member** — the sharpest single lens for this question's domain (see the heuristics below). Owns the non-obvious frame. **Mid model.**
View the complete source on GitHub

Skills included

  • wise-men-flash

Repository facts

License
MIT
Created
Sep 24, 2026
Last pushed
Nov 4, 2026
Forks
0
Open issues
0
Project activity
Active

Before you install

Local scripts

Automated signals, not a guarantee

Checked Nov 4, 2026

Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.

Show 5 notes
  • SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • assets/banner.svg:6 Analysis incomplete: physical or joined line exceeds character limit
  • assets/banner.svg:4 Analysis incomplete: physical or joined line exceeds character limit
  • assets/banner.svg:5 Analysis incomplete: physical or joined line exceeds character limit
Show all 58 signals by file

.claude-plugin/marketplace.json

  • Network accessline 1
    {   "$schema": "https://anthropic.com/claude-code/marketplace.schema.json",
  • Network accessline 2
    "$schema": "https://anthropic.com/claude-code/marketplace.schema.json",
  • Network accessline 3
    "name": "wise-men-flash",   "description": "A three-member Claude council that answers a hard question in about a third of the full wise-men council's time: anchor, one domain lens, Devil's Advocate, one synthesis check, a decision memo.", 
  • Network accessline 4
    "description": "A three-member Claude council that answers a hard question in about a third of the full wise-men council's time: anchor, one domain lens, Devil's Advocate, one synthesis check, a decision memo.",   "owner": { "name": "Ali-ex
  • Network accessline 5
    "owner": { "name": "Ali-expandings", "url": "https://github.com/Ali-expandings" },

.claude-plugin/plugin.json

  • Network accessline 3
    "version": "0.2.0",   "description": "The fast tier of the wise-men council as its own skill: three Claude subagents (a practitioner anchor, one domain lens, a Devil's Advocate) reasoning down different paths, one synthesis check, a decisio
  • Network accessline 4
    "description": "The fast tier of the wise-men council as its own skill: three Claude subagents (a practitioner anchor, one domain lens, a Devil's Advocate) reasoning down different paths, one synthesis check, a decision memo — four calls, a
  • Network accessline 5
    "author": { "name": "Ali-expandings", "url": "https://github.com/Ali-expandings" },
  • Network accessline 6
    "homepage": "https://github.com/Ali-expandings/wise-men-flash",
  • Network accessline 7
    "repository": "https://github.com/Ali-expandings/wise-men-flash",

CHANGELOG.md

  • Network accessline 9
    - **0.1.0** (2026-09-24) — First release: the fast tier of [`wise-men`](https://github.com/Ali-expandings/wise-men) as its own skill. Three members in parallel — a practitioner anchor and a Devil's Advocate on the strong model, one domain m

CONTRIBUTING.md

  • Network accessline 5
    - A protocol change needs evidence — a pre-registered run, or at least a recorded council run — not only reasoning. Say which in `CHANGELOG.md`. - Charts in `assets/` are generated: edit `scripts/make_charts.py`, not the SVGs. - Flash stays
  • Network accessline 6
    - Charts in `assets/` are generated: edit `scripts/make_charts.py`, not the SVGs. - Flash stays four calls. A change that needs peer review or a debate round belongs in the full [`wise-men`](https://github.com/Ali-expandings/wise-men) skill
  • Network accessline 7
    - Flash stays four calls. A change that needs peer review or a debate round belongs in the full [`wise-men`](https://github.com/Ali-expandings/wise-men) skill.

DETAILS.md

  • Network accessline 136
    | | wise-men-flash | [wise-men](https://github.com/Ali-expandings/wise-men) |
  • Network accessline 153
    Against the two rival councils measured in both rounds (features from their READMEs and skill files, as surveyed in the [wise-men landscape](https://github.com/Ali-expandings/wise-men/blob/master/resources/landscape.md), 2026-09-16/17); rou
  • Network accessline 155
    | | wise-men-flash | [Warp council](https://github.com/warpdotdev/common-skills) 24.5k installs | [llm-council](https://github.com/aiwithremy/claude-skills-llm-council) 1.0k installs |

README.md

  • Network accessline 3
    [![stars](https://img.shields.io/github/stars/Ali-expandings/wise-men-flash?style=flat)](https://github.com/Ali-expandings/wise-men-flash/stargazers) [![license](https://img.shields.io/github/license/Ali-expandings/wise-men-flash)](LICENSE)
  • Network accessline 5
    A decision helper for [Claude Code](https://claude.com/claude-code). You ask a hard question, three agents answer it separately (one of them argues against the obvious answer on purpose), and you get back one short memo: what to do, why, th
  • Network accessline 7
    It is the fast version of [wise-men](https://github.com/Ali-expandings/wise-men). About 15 minutes a question instead of 50.
  • Network accessline 42
    Lookups, quick facts, or anything you can undo in a minute: just ask Claude. For big, hard-to-undo decisions, use the full [wise-men](https://github.com/Ali-expandings/wise-men).
  • Network accessline 56
    git clone https://github.com/Ali-expandings/wise-men-flash.git ~/.claude/skills/wise-men-flash
  • Network accessline 60
    **Or with [skills.sh](https://skills.sh):**
  • Network accessline 90
    The fast tier of [wise-men](https://github.com/Ali-expandings/wise-men), extracted into its own skill. Inspired by [karpathy/llm-council](https://github.com/karpathy/llm-council). Design choices drew on — but are **not validated by** — Du e

assets/banner.svg

  • Network accessline 1
    <svg viewBox="0 0 860 190" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round4-axes-dark.svg

  • Network accessline 1
    <svg viewBox="0 0 860 362" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round4-axes.svg

  • Network accessline 1
    <svg viewBox="0 0 860 362" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round4-dark.svg

  • Network accessline 1
    <svg viewBox="0 0 860 456" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round4-questions-dark.svg

  • Network accessline 1
    <svg viewBox="0 0 860 354" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round4-questions.svg

  • Network accessline 1
    <svg viewBox="0 0 860 354" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round4-speed-dark.svg

  • Network accessline 1
    <svg viewBox="0 0 860 472" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round4-speed.svg

  • Network accessline 1
    <svg viewBox="0 0 860 472" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round4.svg

  • Network accessline 1
    <svg viewBox="0 0 860 456" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round5-axes-dark.svg

  • Network accessline 1
    <svg viewBox="0 0 860 464" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round5-axes.svg

  • Network accessline 1
    <svg viewBox="0 0 860 464" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round5-dark.svg

  • Network accessline 1
    <svg viewBox="0 0 860 590" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round5-questions-dark.svg

  • Network accessline 1
    <svg viewBox="0 0 860 450" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round5-questions.svg

  • Network accessline 1
    <svg viewBox="0 0 860 450" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round5-speed-dark.svg

  • Network accessline 1
    <svg viewBox="0 0 860 472" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round5-speed.svg

  • Network accessline 1
    <svg viewBox="0 0 860 472" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

assets/round5.svg

  • Network accessline 1
    <svg viewBox="0 0 860 590" width="860" xmlns="http://www.w3.org/2000/svg" font-family="-apple-system, BlinkMacSystemFont, 'Segoe UI', Helvetica, Arial, sans-serif">

eval-data/README.md

  • Network accessline 5
    **[`head-to-head/` round 5](head-to-head/RESULTS-V5.md)** — this skill's own measurement, run as installed: wise-men-flash 0.1.0 next to wise-men 3.14.0 and every rival from rounds 1–4, on eight new questions, three blind judges each. It is
  • Network accessline 43
    Rounds 1–3 of the study, and the same data seen from the full council: the wise-men repository's [`eval-data/head-to-head/`](https://github.com/Ali-expandings/wise-men/tree/master/eval-data/head-to-head).

eval-data/head-to-head/RESULTS.md

  • Network accessline 54
    `python3 eval-data/head-to-head/flash4.py report` prints the table and comparisons; `results` rewrites this file; `verify` rebuilds every blinded packet from the raw answers and re-parses every judgment, and fails on any difference. Time an

eval-data/head-to-head/flash4.py

  • Executable scriptline 1
    #!/usr/bin/env python3
  • Network accessline 148
    "`python3 eval-data/head-to-head/flash4.py report` prints the table and comparisons; `results` rewrites this file; `verify` rebuilds every blinded packet from the raw answers and re-parses every judgment, and fails on any difference. "     
  • Network accessline 149
    "Time and cost were computed from local agent transcripts when each answer was saved and are stored in the raw file headers; the transcripts themselves are not in the repository. "           "The same data, seen from the full council, is `R
  • Network accessline 150
    "The same data, seen from the full council, is `RESULTS-V4.md` in the [wise-men repository](https://github.com/Ali-expandings/wise-men/blob/master/eval-data/head-to-head/RESULTS-V4.md).", ""]
  • Access to secretsline 152
    [Credential-related source omitted]

eval-data/head-to-head/h2h5.py

  • Executable scriptline 1
    #!/usr/bin/env python3
  • Access to secretsline 256
    [Credential-related source omitted]

scripts/check.sh

  • Executable scriptline 1
    #!/usr/bin/env bash
  • Dangerous shell operationline 55
    stale=$(for f in "$tmp"/*.svg; do cmp -s "$f" "assets/$(basename "$f")" || basename "$f"; done); rm -rf "$tmp"

scripts/make_charts.py

  • Executable scriptline 1
    #!/usr/bin/env python3
  • Access to secretsline 16
    [Credential-related source omitted]
  • Network accessline 33
    r1, sg = F.r1, F.sg def esc(s): return str(s).replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;") def svg(w, h, body, title): return f'<svg viewBox="0 0 {num(w)} {num(h)}" width="{num(w)}" xmlns="http://www.w3.org/2000/svg" font
  • Network accessline 34
    def esc(s): return str(s).replace("&", "&amp;").replace("<", "&lt;").replace(">", "&gt;") def svg(w, h, body, title): return f'<svg viewBox="0 0 {num(w)} {num(h)}" width="{num(w)}" xmlns="http://www.w3.org/2000/svg" font-family="{FONT}">\n<
  • Network accessline 35
    def svg(w, h, body, title): return f'<svg viewBox="0 0 {num(w)} {num(h)}" width="{num(w)}" xmlns="http://www.w3.org/2000/svg" font-family="{FONT}">\n<title>{esc(title)}</title>\n{body}</svg>\n'

How the signals fit together

Facts from a static read of the files at the pinned commit. They describe what the code does, not why.

Some files could not be read; the facts below still hold.

Declared and executed: 1 fact
  • Contacts www.w3.org, which SKILL.md, the README and the plugin manifest do not mentionscripts/make_charts.py:35High confidence
Signal confidence: 31 facts

Works with

Agent compatibility, from the files in this repository
AgentStatusEvidenceInstall to
Claude CodeNativeManifest .claude-plugin/marketplace.json.claude/skills/<skill-name>/
CodexNativeManifest AGENTS.md.agents/skills/<skill-name>/
CursorCompatibleStandard SKILL.md SKILL.md.cursor/skills/<skill-name>/
Gemini CLICompatibleStandard SKILL.md SKILL.md.gemini/skills/<skill-name>/
GitHub CopilotCompatibleStandard SKILL.md SKILL.md.github/skills/<skill-name>/
OpenCodeCompatibleStandard SKILL.md SKILL.md.opencode/skills/<skill-name>/
  • Native The repository ships a manifest, folder or frontmatter field for this agent.
  • Compatible A standard SKILL.md that this agent can read.
  • Unknown No evidence either way.

Derived automatically from paths, manifests and frontmatter. Not tested on each agent.