
Swarm Coordination
- 63 installs
- 111 repo stars
- Updated July 24, 2026
- dralgorhythm/claude-agentic-framework
Helps Claude Code orchestrate multiple sub-agents working in parallel - task decomposition, coordination and result merging.
About
A Claude Code skill for coordinating swarms of sub-agents - decomposing a task, running agents in parallel and merging their results. A solo builder reaches for it when building an agentic system and wanting reliable orchestration of multiple agents rather than a single long-running loop.
- Multi-agent orchestration
- Parallel task decomposition
- Result merging & coordination
Swarm Coordination by the numbers
- 63 all-time installs (skills.sh)
- Ranked #6,243 of 16,546 AI & Agent Building skills by installs in the Skillselion catalog
- Data as of Aug 3, 2026 (Skillselion catalog sync)
npx skills add https://github.com/dralgorhythm/claude-agentic-framework --skill swarm-coordinationAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 63 |
|---|---|
| repo stars | ★ 111 |
| Last updated | July 24, 2026 |
| Repository | dralgorhythm/claude-agentic-framework ↗ |
What it does
Helps Claude Code orchestrate multiple sub-agents working in parallel - task decomposition, coordination and result merging.
Who is it for?
Orchestrating multi-agent swarms
Files
Swarm Coordination
Protocols and patterns for consistent, conflict-free multi-agent development. Follow these guidelines when working alongside other Claude Code agents in the same codebase.
Core Principles
1. Beads as Source of Truth: All work items tracked via bd commands 2. File Locking: Hooks automatically manage file locks - respect them 3. Session Isolation: Each agent has a unique session ID for tracking 4. Clean Handoffs: Always leave state that another agent can continue
File-Based Output
Workers write results to scratchpad/<task-id>.md, not direct context. Only durable artifacts (ADRs, plans, PRDs) go to artifacts/. Orchestrator creates output targets before launching workers; workers write to assigned files; orchestrator reads and synthesizes.
Workflows
Starting Work
- [ ] Check Beads: Run
bd readyto find unblocked issues - [ ] Claim Work: Update issue status:
bd update <id> --status in_progress - [ ] Check Conflicts: Review
.claude/hooks/.file-tracker.logfor recent edits - [ ] Coordinate: If another agent is active, coordinate via Beads comments
During Work
- [ ] Atomic Changes: Make small, complete changes that don't leave broken state
- [ ] Frequent Commits: Commit often to reduce merge conflicts
- [ ] Update Progress: Add comments to Beads issues for visibility
- [ ] Respect Locks: If a file is locked, wait or work on something else
Completing Work
- [ ] Run Tests: Verify changes don't break existing functionality
- [ ] Close Issue:
bd close <id> --reason "Completed: <description>" - [ ] Sync Beads:
bd syncto share updates with other agents - [ ] Clean State: Commit all changes, leave no uncommitted work
Conflict Prevention
File Lock Protocol
Hooks automatically acquire/release locks. If you encounter a lock:
# Check who holds the lock
cat .claude/hooks/.locks/<filename>.lock
# Lock automatically expires after 60 seconds
# If urgent, coordinate via Beads or waitMerge Conflict Strategy
1. Pull frequently: Keep your branch up to date 2. Small PRs: Easier to merge than large changes 3. Coordinate: Use Beads to claim files/features before editing 4. Resolve quickly: Address conflicts immediately when detected
Communication Patterns
Handoff Message
When ending a session with incomplete work:
# Create handoff for next agent
echo '{"message": "Continue implementing auth middleware. Tests passing but needs error handling in src/auth.ts:45"}' > .claude/hooks/.state/handoff.jsonIssue Comments (via Beads)
# Add context for other agents
bd comment <issue-id> "Implemented base class. Needs: validation, tests, docs"Multi-Agent Patterns
Queen-Worker Pattern
For complex tasks, one agent orchestrates while others execute:
1. Queen: Plans, decomposes, assigns via Beads 2. Workers: Claim issues, implement, report completion 3. Sync Point: All workers sync before final integration
Parallel Streams
For independent features:
1. Create separate Beads issues for each stream 2. Each agent claims one stream 3. Avoid editing same files across streams 4. Merge streams at defined integration points
State Files
| File | Purpose |
|---|---|
.claude/hooks/.state/session_*.json | Active agent sessions |
.claude/hooks/.state/handoff.json | Handoff messages between sessions |
.claude/hooks/.locks/*.lock | File edit locks |
.claude/hooks/.file-tracker.log | Recent file modifications |
Best Practices
1. Check Before Edit: Always verify no active locks on target files 2. Complete Units: Finish logical units of work before switching 3. Document Intent: Use Beads issues to declare what you're working on 4. Test Locally: Run tests before pushing to catch issues early 5. Sync Often: Keep Beads and git in sync with other agents
Emergency Procedures
Deadlock Detection
If agents are waiting on each other:
# Check active sessions
ls -la .claude/hooks/.state/session_*.json
# Check active locks
ls -la .claude/hooks/.locks/
# Force release stale locks (use with caution)
find .claude/hooks/.locks -mmin +5 -deleteRecovery from Conflict
1. Save current work to a new branch 2. Sync with main: git fetch && git rebase origin/main 3. Resolve conflicts file by file 4. Update Beads: bd sync 5. Continue work
Integration with Beads
# View all open work
bd list --status open
# Get ready (unblocked) items
bd ready --sort hybrid
# Claim an issue
bd update <id> --status in_progress --assignee claude
# Add dependency
bd dep add <blocking-id> <blocked-id> --type blocks
# Complete work
bd close <id> --reason "Implemented feature X"
# Sync state
bd sync