Skip to content

Hivemind: Claude Code as Orchestrator, opencode as Free Worker Swarm

Engineering - POWERFUL hivemind Source

Install: claude /plugin install engineering-advanced-skills

Claude Code = brain (plans, reviews, merges). opencode = disposable workers on free models (opencode/mimo-v2.5-free default; verified $0.00 per run).

Prerequisites (external dependency)

This skill is a thin orchestration layer over opencode, a third-party CLI. It is not bundled — install and authenticate it yourself first:

Requirement Notes
Node.js >= 18 The scripts use fetch and node:timers/promises.
opencode CLI on PATH npm i -g opencode-ai (or the installer opencode documents).
An authenticated opencode account opencode auth login. Workers run as your account.
Default model opencode/mimo-v2.5-free A free tier offered by opencode, not by Anthropic. Availability, rate limits, and pricing are opencode's to change — override with --model at any time.
Windows only: OPENCODE_GIT_BASH_PATH Point at C:\Program Files\Git\bin\bash.exe, set persistently.

Data flow (read before installing)

Hivemind is opt-in and ships outside every domain bundle. By default your code leaves your machine: the default model, opencode/mimo-v2.5-free, runs on opencode's hosted service, so each worker sends its prompt, plus any files or diffs you give it, to opencode's endpoints. Nothing on the worker side calls the Anthropic API.

To keep code local, point --model at a model served on your own machine (for example an Ollama or LM Studio provider configured in your opencode config). With a hosted model, delegate only code you are allowed to send to that provider, and never delegate secrets.

Setup

  1. Put this skill folder wherever your agent loads skills from (e.g. ~/.claude/skills/hivemind).
  2. Export HIVEMIND_HOME pointing at that folder — the bundled slash commands use it: 
    export HIVEMIND_HOME="$HOME/.claude/skills/hivemind"
  3. Copy the bundled assets into place:
  4. assets/commands/*.md -> ~/.claude/commands/ (the /hive, /oc, /swarm, ... entry points)
  5. assets/agents/*.md -> ~/.config/opencode/agent/ (the scout / coder / tester worker personas)

Both copies are optional: everything the commands do can be driven by invoking scripts/oc-worker.mjs directly, and any opencode agent name works with --agent.

Runtime state (.runs/*.jsonl) is written inside this folder and is gitignored.

Components

Path (relative to this skill dir) Purpose
scripts/oc-worker.mjs ONLY sanctioned way to invoke a worker. Hardened join point.
scripts/oc-status.mjs Fleet progress from run logs (oc-status.mjs <run-id>)
scripts/oc-aggregate.mjs Dedupe/synthesize N worker outputs; consensus findings first
scripts/bench/run-bench.mjs Benchmark configs A (claude solo), B (opencode solo), C (orchestrated swarm)
scripts/bench/grader-prompt.md Blind grading rubric (max 12 pts + PASS/FAIL gate)
scripts/oc-verify.mjs Fan a worker's claims to a verifier on a DIFFERENT model; returns counts + needs_review
assets/commands/ Slash-command entry points to copy into ~/.claude/commands/
assets/agents/ scout / coder / tester / verifier agent definitions for opencode

Slash commands (ship in assets/commands/, copy to ~/.claude/commands/): - /hive <task> - AUTO-ROUTER. Classifies task -> single worker, generic swarm, or template. Default entry point; prefer this over manual routing. - /oc <task> - single worker delegation - /swarm <task> - generic parallel swarm - /review-panel <diff> - 4-lens parallel review (correctness/security/performance/style) + consensus aggregation - /research-sweep <question> - 3-5 parallel research angles, synthesized - /migration <task> - batched per-worktree migration workers + sequenced merge - /test-fleet <target> - partitioned parallel test runs with safety checks

Worker agents (ship in assets/agents/, copy to ~/.config/opencode/agent/): - scout - read-only research (no write/edit/bash) - coder - implements one subtask in its worktree - tester - runs tests only, never edits source

Invocation contract

node "<skill-dir>\scripts\oc-worker.mjs" [--agent scout|coder|tester] [--dir <path>] [--model <p/m>] [--timeout 900] [--run <id> --label <name>] [--json] "TASK TEXT"

Returns exactly ONE compact JSON line: { ok, result, tokens:{total,input,output,cache}, cost_usd, duration_ms, label, agent, model }

On failure: { ok:false, stage:"args"|"exec"|"api"|"parse"|"empty"|"schema", error } with stderr capped at 300 chars.

--json makes result a JSON value: the script extracts it from the reply, gives the model one corrective retry, and fails with stage:"schema" if it still gets no parseable JSON.

--run <id> + --label <name> append lifecycle events (start/done/fail) to .runs/<id>.jsonl inside this skill dir. Use them for EVERY swarm worker so progress is recoverable via oc-status.mjs even after orchestrator context loss.

The script auto-manages the shared server: health-checks 127.0.0.1:4096, spawns opencode serve if dead, waits 5s, falls back to cold start. Workers are idempotent against their --dir; re-run once on ok:false before giving up.

HIVEMIND_SERVER_URL overrides that address (default http://127.0.0.1:4096). It must be a valid URL with a numeric port; anything else fails fast with a single stage:"args" JSON line rather than reaching the spawned process.

Verify instead of re-reading (do this before you trust a claim)

Re-checking every worker claim yourself costs more than the delegation saved. Send the claims to a verifier instead:

node scripts/oc-verify.mjs --dir <same dir the worker saw> --claims findings.json --run <id>

Returns { confirmed, refuted, unsupported, unverified, needs_review[] }. Read needs_review only — those are the refuted and unsupported claims. Confirmed claims are counts, not prose, and do not come back into your context.

HARD RULES for this stage: - The verifier MUST run on a different model from the worker. Same model = same blind spots = a rubber stamp. The verifier defaults to opencode/nemotron-3.5-lightning-free against the worker's default opencode/mimo-v2.5-free; pass --model if you change either. - The verifier sees the FILES and the claims, never the worker's reasoning. - It defaults to REFUTED when uncertain, because a false CONFIRMED is acted on unchecked. - unverified > 0 means the verifier skipped claims. Treat those as unverified, not passed. - Still spot-check a sample yourself. Two free models agreeing is evidence, not proof.

Free models are not always up

Free opencode/* models return provider 404s and empty bodies often enough that an unattended swarm cannot depend on any one of them. Never hardcode a free model id: they stop being free without warning. Pass --model explicitly for scheduled or bulk work, and when a free model fails twice, follow the fallback ladder below rather than retrying.

Golden Rule (non-negotiable)

Raw opencode NDJSON streams must NEVER enter your context. All output arrives via the script's single JSON line. Never pipe opencode run --format json directly into this conversation; never re-implement what the script does.

Single worker flow (/oc)

For one read-only question or small delegation: run oc-worker.mjs without worktrees. Read-only tasks may omit --agent/--dir. Summarize result for the user. If files were written: show git diff before letting the user commit.

Swarm flow (multi-worker)

  1. Decompose task into 2-5 INDEPENDENT subtasks (no shared files).
  2. Writing workers get isolated worktrees FIRST: git worktree add ../<repo>-wt-N -b swarm/N.
  3. Issue ALL worker invocations as PARALLEL Bash tool calls in ONE message.
  4. Review every diff yourself (git diff main...swarm/N). YOU are the only merger.
  5. Merge approved branches, remove worktrees, run tests.
  6. Report table: subtask | agent | tokens | outcome + total worker tokens.

HARD RULES: workers never share directories; never delegate merging/reviewing; escalate to your own Sonnet only when a free-model worker demonstrably fails twice.

Benchmarking

node scripts\bench\run-bench.mjs --repo <project> [--configs a,b,c] [--task 1-5]
Appends JSONL records (ts, config, tokens, cost, duration) to bench-results.jsonl. Grade artifacts blind with grader-prompt.md (grader sees only task spec + output). Configs: A=claude solo baseline, B=opencode solo, C=claude orchestrating 2 workers.

Fallback ladder (all flows)

  1. Worker ok:false -> re-invoke once against the same dir.
  2. Still failing -> orchestrator performs that subtask inline, marks it [orchestrator-sourced].
  3. opencode entirely down (exec/api twice) -> announce, abandon workers, do the task directly. Never let a swarm fail a task that Claude could have done itself.

Fleet patterns

Four reusable topologies ship as slash commands (see table above). Shared invariants: parallel spawns in one message; --run/--label on every worker; aggregation via oc-aggregate.mjs when 3+ workers produce findings; consensus beats single-lens claims; worktree isolation whenever any worker writes.

Windows notes (hard-won)

  • Requires OPENCODE_GIT_BASH_PATH=C:\Program Files\Git\bin\bash.exe (set persistently).
  • The script resolves the REAL opencode.exe by parsing the npm .cmd shim — Node's EINVAL policy blocks spawning .cmd directly. Do not "simplify" resolver back to where.exe first-line.
  • Free models: opencode/mimo-v2.5-free, opencode/nemotron-3.5-lightning-free, opencode/hy3-free. NOTE: opencode-go/* models require workspace billing — avoid.

Known limits

  • Free-tier rate limits can 429 under heavy swarms; space out retries.
  • Worker quality varies; always review diffs. Scout answers are evidence-cited.
  • Bench config C consumes real Claude tokens for orchestration (~1-2k/task).

Anti-patterns

Anti-pattern Why it breaks Do this instead
Piping opencode run --format json straight into the orchestrator Raw NDJSON floods context — the exact cost the skill exists to avoid Always go through scripts/oc-worker.mjs, which returns one compact JSON line
Two writing workers in one directory Concurrent edits corrupt each other's diffs One git worktree per writing worker, created before the spawn
Letting a worker merge, review, or approve its own branch Free-tier workers are the least reliable judges of their own output The orchestrator is the only merger and the only reviewer
Spawning workers sequentially, one per message Loses the entire wall-clock benefit of a swarm Issue every worker invocation as parallel calls in ONE message
Retrying a failing worker indefinitely Burns rate limit and stalls the task Retry once, then do the subtask inline and mark it [orchestrator-sourced]
Delegating secrets, credentials, or private code Worker traffic leaves for opencode's endpoints Keep sensitive context in the orchestrator; send workers only what is safe to share
Trusting cost_usd: 0 as a permanent guarantee The free tier belongs to opencode and can change Re-check pricing before relying on zero cost for bulk work

Cross-references

  • engineering/llm-cost-optimizer — decide whether a task is worth delegating before Hivemind decides how
  • engineering/agent-harness — harness patterns for the orchestrator side of the loop
  • engineering/workflow-builder — for deterministic pipelines that do not need independent worker judgment
  • engineering/skills and engineering/write-a-skill — authoring conventions used by the worker agent definitions in assets/agents/