Skip to content

diegosouzapw/awesome-omni-skill / skills/ai-agents/agent-coordination

agent-coordination

Verified

Agent assignment matrix, blocker escalation, and TDM coordination patterns. Use when assigning work to specialists, managing blockers, or coordinating multi-agent workflows.

Agent Coordination Skill

Purpose

Guide correct agent assignment, blocker escalation, and delivery coordination following the TDM role boundaries.

When This Skill Applies

Invoke this skill when:

  • Assigning work to specialist agents
  • Managing blockers across agents
  • Coordinating multi-agent workflows
  • Escalating issues to ARCHitect/POPM
  • Verifying pre-implementation gate
  • Updating Linear with evidence

Stop-the-Line Conditions

FORBIDDEN Patterns

# FORBIDDEN: Wrong agent assignment
Database work → BE Developer    # WRONG: Use Data Engineer!
Security validation → QAS       # WRONG: Use Security Engineer!
Documentation → BE Developer    # WRONG: Use Tech Writer!

# FORBIDDEN: Skipping pre-implementation gate
Start coding before BSA spec    # Must have acceptance criteria first

# FORBIDDEN: Unresolved blockers >4 hours
Blocker exists but not escalated  # Escalate within 1 hour

CORRECT Patterns

# CORRECT: Right agent for the job
Database work → Data Engineer
Security validation → Security Engineer
Documentation → Tech Writer
Planning/Specs → BSA
Architecture review → System Architect
API implementation → BE Developer
UI implementation → FE Developer
Testing → QAS
PR/Release → RTE

# CORRECT: Pre-implementation gate
1. BSA creates spec with acceptance criteria
2. System Architect reviews patterns
3. THEN specialist executes

Agent Assignment Matrix (MANDATORY)

Work TypeCorrect AgentNever Use
Database/MigrationsData EngineerBE Developer
Security/RLSSecurity EngineerQAS
DocumentationTech WriterBE/FE Developer
Specs/PlanningBSAAny implementation
ArchitectureSystem ArchitectDirect to developer
API RoutesBE DeveloperFE Developer
UI ComponentsFE DeveloperBE Developer
Testing/QAQASImplementation team
PR/ReleasesRTEDevelopers

Reference: docs/workflow/TDM_AGENT_ASSIGNMENT_MATRIX.md

Blocker Escalation Protocol

When to Escalate

ConditionEscalate ToDeadline
Blocker > 1 hourTDMImmediately
Blocker > 4 hoursARCHitectUrgent
Architecture ambiguityARCHitectBefore work
Cross-team dependencyTDM + POPMSame day
Security concernSecEngImmediately

Escalation Template

**Blocker Escalation**

**Ticket**: {TICKET_PREFIX}-XXX
**Blocked Since**: [timestamp]
**Agent**: [which specialist is blocked]

**Issue**:
[Clear description of what's blocking progress]

**Attempts Made**:

1. [What was tried]
2. [What was tried]

**Request**:
[Specific ask - decision needed, resource needed, etc.]

Pre-Implementation Gate

MANDATORY before any implementation work:

1. BSA creates spec with:
   - [ ] Clear acceptance criteria
   - [ ] Pattern references for execution
   - [ ] Success validation command

2. System Architect reviews:
   - [ ] Pattern compliance
   - [ ] RLS requirements (if database)
   - [ ] Security implications

3. THEN implementation can begin

Evidence Attachment

All work must include evidence in Linear:

**Implementation Evidence**

**Session ID**: [Claude session ID]
**Agent**: [which specialist]
**Ticket**: {TICKET_PREFIX}-XXX

**Work Completed**:

- [x] Task 1
- [x] Task 2

**Validation**:

```bash
yarn ci:validate
# All checks passed
```

Next Steps: [if any]


## TDM Boundaries

### TDM Does

- ✅ Track delivery progress
- ✅ Update Linear tickets
- ✅ Escalate blockers
- ✅ Attach evidence
- ✅ React to issues

### TDM Does NOT

- ❌ Orchestrate feature work (ARCHitect's job)
- ❌ Run CI/CD validation (specialists' job)
- ❌ Execute technical work (specialists' job)
- ❌ Proactively assign features (ARCHitect's job)

## Authoritative References

- **Agent Assignment Matrix**: `docs/workflow/TDM_AGENT_ASSIGNMENT_MATRIX.md`
- **Agent Workflow SOP v1.3**: `docs/sop/AGENT_WORKFLOW_SOP.md`
- **CONTRIBUTING.md**: Workflow requirements
- **Linear SOP skill**: Evidence attachment templates
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 468kUpdated Claude CodeCodexUnverifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 319kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 319kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 319kUpdated Claude CodeCodexVerifiedInstall