Opencode Swarm

Unclaimed

by ZaxbyHub

Architect-centric agentic swarm plugin for OpenCode. Hub-and-spoke orchestration with SME consultation, code generation, and QA review.

Install

$bunx opencode-swarm install

Unclaimed listing

Is this your MCP server?

This listing was auto-indexed from the public record. Claim it to edit the page, set compatibility and unlock growth tools. Takes under two minutes.

Claim this server

Tools (31)

build_check

Runs your project's native build/typecheck

complete_pr_workflow

Architect-only: validates terminal PR-review or PR-feedback coverage; feedback uses a two-call arm/publish protocol and clears only after the approved content is observed on both the bound remote-tracking ref and the actual remote branch

diff

Structured git diff with contract change detection; supports summaryOnly mode returning file list with additions/deletions counts

external_skill_delete

Remove candidate from quarantine store

external_skill_discover

Fetch and validate a skill from a configured source

external_skill_inspect

View full candidate details

external_skill_list

List candidates with status filters

external_skill_promote

Promote validated candidate to active skill (user approval required)

external_skill_reject

Reject candidate with reason

external_skill_revoke

Retire a previously promoted skill

generate_mutants

Architect-only: generates LLM-based mutation patches (5–10 per function across 6 types: off-by-one, null substitution, operator swap, guard removal, branch swap, side-effect deletion) for direct consumption by the mutationtest tool; returns SKIP verdict on LLM failure rather than throwing

git_blame

Per-line git blame metadata (sha, author, date, summary) via git blame --porcelain; supports optional line range filtering

incremental_verify

Post-coder hook for TS/JS, Go, Rust, Python, and C#; configured by incrementalverify., not invoked as a registered tool

max_lines

Default line limit for any tool output.

mutation_test

Applies LLM-generated mutation patches to source files and runs tests to measure kill rate; verdict is pass/warn/fail based on configurable thresholds; used by the mutationtest gate (opt-in, off by default)

parse_lane_candidates

Architect-only: parses [CANDIDATE] rows and micro-lane [CLEAN] attestations from a lane outputref; supports asserted batch families, preserves dispatch provenance, and persists sidecar JSONL records

per_tool

Overrides maxlines for specific tools. The diff and symbols tools are truncated by default because their outputs can be very large.

phase_complete

Enforces phase completion, verifies required agents, requires a valid retrospective evidence bundle, logs events, and resets state; appends to events.jsonl with file locking

placeholder_scan

Catches TODOs, FIXMEs, stubs, placeholder text

pre_check_batch

Runs lint, secretscan, SAST, and quality budget in parallel (~15s vs ~60s sequential)

quality_budget

Enforces complexity, duplication, and test ratio limits

run_pr_feedback_stage_a

Architect-only: executes targeted reproduction/regression and exact git diff --check plus every mechanically applicable workspace/category/source build, typecheck, and lint obligation, then persists content-bound Stage A receipts before ordered feedback review gates

sast_scan

Offline security analysis, 68 rules, 8 languages

sbom_generate

CycloneDX dependency tracking, 8 ecosystems

suggest_patch

Reviewer-safe structured patch suggestion; supports format parameter ('json' or 'unified') where unified outputs valid unified diff with diff --git headers, hunks, and context

symbols

Extract exported symbols from source files; supports workspace (boolean) and name (string) parameters for multi-file symbol search

syntax_check

Tree-sitter validation across the configured language grammar map

test_runner

Auto-detect and run tests; supports bail parameter to inject framework-specific bail flags for early exit on first failure

truncation_enabled

Global switch (default true).

write_mutation_evidence

Architect-only: writes mutation gate results atomically to .swarm/evidence/{phase}/mutation-gate.json; accepts verdict (PASS/WARN/FAIL/SKIP), kill rate metrics, and optional survived mutant details; normalizes uppercase-to-lowercase before persisting

write_pr_review_trigger_eval

Architect-only: validates the exact mandatory PR-review micro-lane ID set, requires completed provenance for all 11 repository-agnostic lanes, verifies the claimed merge base against an exact live base ref, rejects every NO-MATCH waiver, and atomically persists .swarm/pr-review/<runid>/trigger-eval.