Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/n3m6-swarm
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: documents, testing, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions: not checked yet.
What's inside
- Skill name
swarm- Description
- Fan out N parallel workers, drain them, and return one report. Use for @swarm, 'swarm this', or parallel coverage, races, gauntlets, and exploration.
- License
- GPL-3.0
- Metadata
- opencode/autoinvoke: false
- Instructions
- 12 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Swarm Run coverage slices or independent attempts and return one consolidated report. 1. Frame the goal, scope, done predicate, and required artifact. Choose partitioned coverage, a race, or a mix. For a race, declare `first pass`, `rank all`, or `best-of` before launch. N is the total work count, not the concurrency limit. 2. Read [OpenCode runtime](../../references/opencode-runtime.md). Use the configured `swarm-worker` role, inherited by default. Match concurrency to the actual session and `max_workers`. Give each writer a separate checkout or disjoint writable scope. A verifier's brief names exact commits and measurement methods when applicable. 3. Launch permitted workers with complete briefs: goal, slice or arm, artifact path, verification commands, and reporting contract. Queue excess work. If delegation is prohibited or unavailable, cover the slices directly and record that independence or parallelism was unavailable. 4. Each worker returns `PASS`, `ISSUES`, or `BLOCKED` with evidence. A proof of a defect is `ISSUES`; list all proven issues in that slice. A missing required slice is a gap. For malformed reports, request the missing evidence once, then record the gap if it remains. 5. Verify the returned artifacts and apply the declared coverage or race rule. Same-model runs are independent attempts; model or provider diversity must be evidenced by what actually ran. 6. Return a compact results table, deduplicated issues, gaps or dropouts, and the selection rule. Preserve artifact paths without pasting raw worker dumps.
From n3m6/opencode-config, GPL-3.0.
More from n3m6/opencode-config
- 0.640score
unslop
n3m6
Edit prose to remove AI writing patterns while preserving meaning, intended tone, and the requested format.
SkillAFreshness: A★ 19Claude Code +9 - 0.635score
no-comments
n3m6
Review comments and workaround explanations in a scoped diff, preserve necessary contracts, and fix accepted findings.
SkillAFreshness: A★ 19Claude Code +9 - 0.635score
setup-pstack
n3m6
Configure which models pstack uses per role and at what reasoning budget.
SkillAFreshness: A★ 19Claude Code +9 - 0.620score
architect
n3m6
Sketch types, signatures, and module structure before code, then stay in the loop while implementation fills in.
SkillAFreshness: A★ 19Claude Code +9 - 0.620score
create-verification-skill
n3m6
Generate a project-local verification skill that drives your app the way a user does — any language, framework, or platform.
SkillAFreshness: A★ 19Claude Code +9 - 0.620score
how
n3m6
Use for "how does X work", code walkthroughs before changing something, and placement / ownership / layering questions ("where should this live", "which…
SkillAFreshness: A★ 19Claude Code +9 - 0.620score
interrogate
n3m6
Use for "interrogate", "adversarial review", "multi-model review", "challenge this", "stress test this code", "find blind spots", or "tear this apart".
SkillAFreshness: A★ 19Claude Code +9 - 0.620score
poteto-mode
n3m6
poteto's agent style for concise, detailed responses, deliberate subagents, unslopped prose, simple code, and verified work.
SkillAFreshness: A★ 19Claude Code +9 - 0.618score
figure-it-out
n3m6
Design an auditable playbook when no narrower one fits: a large migration, an ambitious multi-part change, or work a human reviews after stepping away.
SkillAFreshness: A★ 19Claude Code +9 - 0.618score
benchmark-checklist
n3m6
Vet a perf measurement (limiter, tuning, limits, errors, repeatability, relevance, and whether the work happened) before you report or act on it.
SkillAFreshness: A★ 19Claude Code +9 - 0.618score
correct
n3m6
Find the mistakes agents keep repeating in this repo and make each one impossible.
SkillAFreshness: A★ 19Claude Code +9 - 0.615score
maintain-verification-skill
n3m6
Periodic pass that keeps a project's verification skill and feature map honest: parallel source readers per feature, one live session driving every feature…
SkillAFreshness: A★ 19Claude Code +9
Signals
- Stars
- 19
- Forks
- 2
- Open issues
- 1
- Last commit (this folder)
- Oct 6, 2026 (22 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- GPL-3.0
- Repository
- github.com/n3m6/opencode-config
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Same skill elsewhere
Copies of this SKILL.md published by others; search shows only one of them.
- 0.761score
swarm
backnotprop
Fan out N parallel workers, drain them, and return one report.
SkillAFreshness: A★ 2.7KClaude Code +9 - 0.406score
swarm
uzairansaruzi
Fan out N parallel workers, drain them, and return one report.
SkillAFreshness: A★ 155Claude Code +9 - 0.393score
swarm
minghinmatthewlam
Fan out N parallel workers, drain them, and return one report.
SkillAFreshness: A★ 46Claude Code +9 - 0.385score
swarm
jiangkoumo
Fan out N parallel workers, drain them, and return one report.
SkillAFreshness: A★ 27Claude Code +9 - 0.384score
swarm
iap
Fan out N parallel workers, drain them, and return one report.
SkillAFreshness: A★ 27Claude Code +9 - 0.375score
swarm
peterpme
Fan out N parallel workers, drain them, and return one report.
SkillBFreshness: B★ 93Claude Code +9 - 0.367score
swarm
shrimpwtf
Fan out N parallel workers, drain them, and return one report.
SkillBFreshness: B★ 39Claude Code +9 - 0.333score
swarm
olibyte
Fan out N parallel workers, drain them, and return one report.
SkillBFreshness: B★ 2Claude Code +9 - 0.323score
swarm
dumkydewilde
Fan out N parallel workers, drain them, and return one report.
SkillBFreshness: B★ 1Claude Code +9
More from n3m6
- 0.640score
unslop
n3m6
Edit prose to remove AI writing patterns while preserving meaning, intended tone, and the requested format.
SkillAFreshness: A★ 19Claude Code +9 - 0.635score
setup-pstack
n3m6
Configure which models pstack uses per role and at what reasoning budget.
SkillAFreshness: A★ 19Claude Code +9 - 0.635score
no-comments
n3m6
Review comments and workaround explanations in a scoped diff, preserve necessary contracts, and fix accepted findings.
SkillAFreshness: A★ 19Claude Code +9 - 0.620score
interrogate
n3m6
Use for "interrogate", "adversarial review", "multi-model review", "challenge this", "stress test this code", "find blind spots", or "tear this apart".
SkillAFreshness: A★ 19Claude Code +9 - 0.620score
create-verification-skill
n3m6
Generate a project-local verification skill that drives your app the way a user does — any language, framework, or platform.
SkillAFreshness: A★ 19Claude Code +9 - 0.620score
how
n3m6
Use for "how does X work", code walkthroughs before changing something, and placement / ownership / layering questions ("where should this live", "which…
SkillAFreshness: A★ 19Claude Code +9
Cite this page
swarm skill. Agent Mart. https://agentmart.dev/skills/n3m6-swarm (last updated 2026-10-24)