swarm-implement · zaxbyhub, claude — Claude Code skill: what it does, how to install
The swarm-implement skill for Claude Code takes a feature or bug fix from exploration through validation, following the repo's canonical SKILL.md.
MIT ★ 539 updated 2026-10-29
npx skills add ZaxbyHub/opencode-swarm --skill swarm-implement -a claude-codeWhat swarm-implement does
The skill body barely describes anything. It points at ../../../.opencode/skills/swarm-implement/SKILL.md and tells the agent to follow that file as the workflow. So it earns its keep inside a checkout of ZaxbyHub/opencode-swarm, where the canonical file sits at that relative path. Copy it into an unrelated project and the pointer goes nowhere.
The Claude notes carry the execution rules. Exploration of disjoint areas and independent review go to parallel subagents. Validation uses whatever commands the repository already relies on. Reviewer and critic approval must land in the task artifacts before the final synthesis, and explorer notes, a green test run or a self-check do not count as that approval while subagents are available.
One more thing in the frontmatter: disable-model-invocation: true. The model will not reach for this skill unprompted, so you invoke it yourself.
What it helps with
- Carry a feature through exploration, implementation and review without dropping a stage
- Hand exploration and independent review to parallel subagents
- Write reviewer approval into the task artifacts before synthesis
- Stop a passing test run from standing in for the review gate
- Choose validation commands that match the repository
Example prompts for the agent
- Take
swarm-implementand refactor the billing module: exploration first, then implementation with a subagent review. - Run the three-file bug fix through
swarm-implementand validate with the commands the repo already uses. - Use
swarm-implementon this refactor and record the reviewer approval in the task artifacts.
How to install swarm-implement
1Skills CLI (Vercel)
npx skills add ZaxbyHub/opencode-swarm --skill swarm-implement -a claude-code2Manually: put the files in place
~/.claude/skills/swarm-implement/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/swarm-implement/
1Skills CLI (Vercel)
npx skills add ZaxbyHub/opencode-swarm --skill swarm-implement -a codex2Manually: put the files in place
.agents/skills/swarm-implement/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add ZaxbyHub/opencode-swarm --skill swarm-implement -a cursor2Manually: put the files in place
.cursor/skills/swarm-implement/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install ZaxbyHub/opencode-swarm .claude/skills/swarm-implement2Manually: put the files in place
.github/skills/swarm-implement/SKILL.md1Skills CLI (Vercel)
npx skills add ZaxbyHub/opencode-swarm --skill swarm-implement -a gemini-cli2Platform CLI
gemini skills install https://github.com/ZaxbyHub/opencode-swarm --path .claude/skills/swarm-implementManually: put the files in place
.devin/skills/swarm-implement/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add ZaxbyHub/opencode-swarm --skill swarm-implement -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-implement
How do I install swarm-implement in Claude Code?
Run npx skills add ZaxbyHub/opencode-swarm --skill swarm-implement -a claude-code. The skill file lands at ~/.claude/skills/swarm-implement/SKILL.md.
What does swarm-implement do?
It is the entry point to the opencode-swarm workflow: the agent opens the canonical SKILL.md from the repository and works from it. The Claude version adds rules on subagent delegation, validation commands and recording approvals.
Why doesn't the skill fire on its own?
Its frontmatter sets disable-model-invocation: true, so the model will not pick it up by itself. You have to call it.
Source text
SKILL.md · MIT
--- name: swarm-implement description: Claude Code adapter for the canonical opencode-swarm implementation workflow. Use for feature work, bug fixes, refactors, and multi-file changes that need exploration, implementation, review, and validation. disable-model-invocation: true metadata: audience: swarm-plugin --- # Swarm Implement Read and follow `../../../.opencode/skills/swarm-implement/SKILL.md` as the canonical workflow. ## Claude Code Execution Notes - Use parallel subagents for disjoint exploration and independent review. - Use objective validation commands appropriate to the repository. - For changed-work tasks, record reviewer and critic approvals in durable task artifacts before final synthesis. - Do not count explorer output, passing tests, or self-review as the final implementation reviewer gate when subagent delegation is available.
© ZaxbyHub · MIT · ZaxbyHub/opencode-swarm
Similar skills
All skills for Agent setup: memory, context, orchestration →swarm-implement by other authors comes first.