swarm · zaxbyhub — Claude Code skill: what it does, how to install
Enables a rigorous swarm-style posture with parallel searches, visible plan updates, and dual review gates for code changes.
MIT ★ 539 updated 2026-10-29
npx skills add ZaxbyHub/opencode-swarm --skill swarm -a claude-codeWhat swarm does
The skill adapts swarm behavior patterns for Codex and Claude Code sessions without calling external OpenCode commands. The agent runs parallel local searches where safe, keeps a visible task checklist using update_plan, and provides brief status updates during lengthy operations.
Every change to code, tests, docs, or package metadata must pass strict exit criteria. Final sign-off requires objective validation data, reviewer approval on the latest diff, and separate sign-off from a critic. Edits after these approvals are strictly forbidden. Passing tests or self-review do not count as reviewer approval when subagent delegation is authorized.
Verdicts stay documented in repository files. Tasks tracking bugs use 08b-implementation-review.md and 09-final-critic.md, while other work uses task-local artifacts. The skill also recommends commands such as $swarm-implement and $swarm-pr-review for dedicated phases.
What it helps with
- run parallel code searches across local repository files
- track complex refactoring progress through
update_plan - enforce two-tier reviews by a distinct reviewer and a critic
- record approval logs in artifacts like
08b-implementation-review.md
Example prompts for the agent
- Run in swarm mode and inspect the authentication module
- Refactor this handler using
update_planand prepare critic artifacts - Implement the pending bug fix following the
$swarm-implementposture
How to install swarm
1Skills CLI (Vercel)
npx skills add ZaxbyHub/opencode-swarm --skill swarm -a claude-code2Manually: put the files in place
~/.claude/skills/swarm/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/swarm/
1Skills CLI (Vercel)
npx skills add ZaxbyHub/opencode-swarm --skill swarm -a codex2Manually: put the files in place
.agents/skills/swarm/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add ZaxbyHub/opencode-swarm --skill swarm -a cursor2Manually: put the files in place
.cursor/skills/swarm/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install ZaxbyHub/opencode-swarm .agents/skills/swarm2Manually: put the files in place
.github/skills/swarm/SKILL.md1Skills CLI (Vercel)
npx skills add ZaxbyHub/opencode-swarm --skill swarm -a gemini-cli2Platform CLI
gemini skills install https://github.com/ZaxbyHub/opencode-swarm --path .agents/skills/swarmManually: put the files in place
.devin/skills/swarm/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add ZaxbyHub/opencode-swarm --skill swarm -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about swarm
How to install swarm in Claude Code?
Run npx skills add ZaxbyHub/opencode-swarm --skill swarm -a claude-code in your terminal. The configuration installs to ~/.claude/skills/swarm/SKILL.md.
What does the swarm skill do?
It applies a high-rigor swarm workflow that enforces parallel exploration, transparent step tracking, and double-review approvals before accepting code changes.
What gates are required to complete code edits?
You need objective validation proof, reviewer approval on the exact diff, subsequent critic approval, zero unresolved blockers, and no code edits after sign-off.
Source text
SKILL.md · MIT
--- name: swarm description: Codex adapter for enabling a higher-rigor swarm-like workflow in the current Codex thread. Use when the user asks for swarm mode, maximum quality, parallel investigation, extra review rigor, or a swarm-style approach without necessarily invoking the OpenCode Swarm plugin. metadata: audience: swarm-plugin --- # Swarm Read `.opencode/skills/swarm/SKILL.md` for the source behavior model, then adapt it to Codex. Codex-specific execution notes: - Treat this as a workflow posture, not an instruction to call OpenCode `/swarm` commands. - Use parallel local reads/searches where safe. - Use `update_plan` for visible task tracking on substantial work. - Add independent review pressure through `$qa-sweep`, `$swarm-pr-review`, or self-critic checks when subagents are unavailable. - Keep the user looped in with short progress updates during longer work. - For any task that edits code, tests, docs, package metadata, release notes, or skill files, final completion requires: 1. objective validation evidence, 2. independent implementation reviewer approval on the actual latest diff, 3. separate critic approval after reviewer approval, 4. no unresolved `NEEDS_REVISION`/`BLOCKED` items, and 5. no edits after those approvals. - Record reviewer and critic verdicts in durable task artifacts. For issue-tracer work, use `08b-implementation-review.md` and `09-final-critic.md`; otherwise use task-local review artifacts unless the repo forbids artifacts. - Explorer subagents, a plan critic, passing tests, or self-review do not satisfy the final implementation-review gate when subagent delegation is available and swarm work is authorized. For implementation tasks, prefer `$swarm-implement`; for fresh PR review, prefer `$swarm-pr-review`; for known PR feedback closure, prefer `$swarm-pr-feedback`; for bug tracing, prefer `$issue-tracer`.
© ZaxbyHub · MIT · ZaxbyHub/opencode-swarm
Similar skills
All skills for Agent setup: memory, context, orchestration →swarm by other authors comes first.