메뉴
Meta-agent orchestrator for complex multi-step projects. Routes tasks to specialized agents via DAG decomposition, resolves conflicting agent outputs with evidence-based arbitration, provides cost-aware delegation, and includes RF/ham radio calculation reference (antenna, link budget, coax, SWR, AREDN).
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Continue iterative work from the last ACCEPTED result plus only the new requested change, instead of replaying the conversation. Use during long multi-turn editing of a document, config, or code file; when the user says "keep going from what we had", asks for a follow-up change to something already approved, or is near plan/token limits in an iterative session. Saves one integrity-checked accepted result, builds a minimal continuation prompt, and never carries rejected drafts or transcript history forward.
Use this when: my tests keep failing randomly, write tests for this code, tests pass locally but fail in CI, how do I mock an HTTP call, my test suite is too slow, debug a failing test, increase test coverage, how do I test a database query, write e2e tests for a user journey, run load tests against my API, set up a test framework, add TDD to my workflow, pytest, Vitest, evaluate my agent, LLM eval, agent flaky, RAG eval, prompt regression, hallucination test, Pass@N, LLM-as-Judge, trace agent execution, test my LangGraph agent, agent guardrails, eval dataset, mock agent tools, agent cost ceiling
Use this when: my API keeps returning 401 or 403, getting 429 rate limit errors, integrate a third-party API, my requests keep timing out, add retry logic with backoff, design a REST API, handle OAuth2 authentication, paginate API responses, CORS errors blocking my requests, secure API keys properly, build consistent error responses, my webhook isn't receiving events, OpenAPI, JWT, circuit breaker, idempotency key, API versioning, rate limiting, token refresh
Use this when: my pipeline is failing, set up CI, speed up my build, why is CI slow, cache miss every build, automate releases, set up GitHub Actions, self-hosted runner not connecting, fix my build, run jobs in parallel, deploy on every merge, build and push Docker image, my workflow is flaky, release automation, test pipeline locally before pushing, OIDC auth in CI, semantic versioning
Use this when: review my code, what's wrong with this function, find bugs before they hit production, is this code secure, check for SQL injection, my error handling is wrong, find edge cases I missed, improve this messy code, this PR needs a review, catch performance problems, my code might have a security vulnerability, refactor this for readability, is this logic doing the right thing, N+1 query, race condition, resource leak, mutable default args, unhandled promise
Use this when: build a data pipeline, my pipeline is not idempotent, clean messy data, convert CSV to Parquet, my data has duplicates, validate schema at ingestion, pipeline fails on re-run, process files larger than memory, schedule a recurring job, upstream schema changed and broke my pipeline, migrate data between systems, query Parquet without loading it, deduplicate records, batch vs stream processing, DuckDB for analytics, choose an orchestrator, slow pandas pipeline
| name | mission-control-pro |
| description | Meta-agent orchestrator for complex multi-step projects. Routes tasks to specialized agents via DAG decomposition, resolves conflicting agent outputs with evidence-based arbitration, provides cost-aware delegation, and includes RF/ham radio calculation reference (antenna, link budget, coax, SWR, AREDN). |
| triggers | ["delegate this across agents","orchestrate multiple tasks","break this down into parallel work","coordinate these systems","RF calculation needed","antenna design","ham radio link budget","resolve disagreement from parallel agents","show progress on complex project"] |
| tags | ["orchestration","multi-agent","reasoning","RF","ham-radio","dag","conflict-resolution"] |
| author | next-gen |
Executive layer above isolated agents. Decompose complex tasks into DAGs, route to specialists, resolve conflicts with evidence scoring, maintain audit trails. For RF/ham radio: prefer local formula over API call (cost = $0, instant).
| Domain | Default | Notes |
|---|---|---|
| Task decomposition | DAG | Identify parallel vs. sequential subtasks |
| Conflict resolution | Evidence scoring | Weighted criteria, not majority vote |
| Cost awareness | Local-first | Formulas/local logic before external APIs |
| State tracking | Task registry | state + agent + deps + acceptance criteria |
| RF calculator | Built-in formulas | Dipole, vertical, Yagi, coax, SWR, link budget |
IF task involves multiple parallel workstreams:
→ Decompose to DAG: identify independent subtasks → assign agents → aggregate
→ Track each subtask: state(In-Progress/Blocked/Completed), agent, deps
IF parallel agents produce conflicting outputs:
→ Evidence scoring: weight criteria against user context
→ Score each option: sum weighted criteria
→ Recommend highest scorer; explain trade-offs
→ If agents loop A→B→A→B twice: HALT, escalate to human
IF choosing between tools/databases:
→ Check local decision matrix first (cost $0, instant)
→ Only escalate to expensive API if local confidence < 0.70
IF RF/antenna calculation needed:
→ Use built-in formulas (see RF Reference below)
→ Always show: formula, inputs, result, operational meaning
IF debugging multi-agent failure:
→ RCA: Observation → History → Hypothesis(×3) → Isolation test
| Anti-Pattern | Use Instead |
|---|---|
| Sequential execution when parallel is possible | DAG decomposition |
| Accepting first agent output without conflict check | Evidence-based arbitration |
| Calling expensive API for deterministic calculation | Local formula/decision matrix |
| A→B→A→B delegation loop | HALT at 2 cycles, escalate to human |
| Black-box recommendations | Show scoring + evidence always |
| No audit trail | Structured audit log per task |
ROUTER_KEYWORDS = {
"orchestrator": ["delegate", "coordinate", "parallel", "DAG", "checkpoint"],
"reasoner": ["why", "decision", "root cause", "trade-off", "analysis"],
"rf_calculator": ["antenna", "RF", "coax", "SWR", "link budget", "repeater", "DMR"],
"tech_advisor": ["database", "framework", "architecture", "choose between"],
"conflict_resolver": ["contradiction", "disagree", "conflicting", "merge results"]
}
def route_task(prompt: str) -> list[str]:
matched = [agent for agent, kws in ROUTER_KEYWORDS.items()
if any(kw.lower() in prompt.lower() for kw in kws)]
return matched or ["orchestrator"]
4-step process:
1. Identify subtasks — parse implicit parallel work
2. Map dependencies — which tasks block others?
3. Assign agents — route each to specialist
4. Aggregate — merge outputs, detect conflicts
Example — "Deploy RF system with AREDN mesh":
├── [PARALLEL] RF analysis → antenna design → AREDN topology
│ └─ Feeds: link budget validation
├── [PARALLEL] Database selection → VLAN design
│ └─ Feeds: integration test plan
├── [SEQUENTIAL] Conflict resolution
└── [SEQUENTIAL LAST] Execute with fallback
User context: "RF monitoring system, 100ms latency requirement"
Agent A: "Use Postgres (ACID, structured)"
Agent B: "Use Redis (fast, real-time)"
Criteria Weight Postgres Redis
Durability 0.4 1.0→0.40 0.3→0.12
Latency 0.4 0.6→0.24 0.9→0.36
Schema 0.2 1.0→0.20 0.2→0.04
Total: 0.84 0.52
Resolution: Postgres. Consider Redis cache if p95 > 100ms under load.
| DB | Score | Best For |
|---|---|---|
| TimescaleDB | 0.84 | RF/IoT time-series |
| PostgreSQL | 0.82 | Structured, relational |
| Redis | 0.66 | Cache layer, ephemeral |
| MongoDB | 0.66 | Flexible schema |
audit_log:
task_id: deploy-rf-system
decomposition:
- subtask: antenna_design
agent: rf_calculator
status: complete
result: "97.7cm dipole, 50Ω"
conflicts:
- contradiction: frequency offset
resolution: "Evidence scoring: +600kHz (FM standard)"
checkpoints:
- "Antenna validated against SWR model"
- "All subtasks complete"
Length (m) = 142.65 / f_MHz
146 MHz → 0.977m (97.7cm) WHY: Half-λ resonance ≈ 50Ω, 2.15 dBi
Length (m) = 71.33 / f_MHz
146 MHz → 0.489m (48.9cm) WHY: Ground plane as image, omni pattern
Gain (dBi) ≈ 10 + 8.5 × log10(f_MHz)
146 MHz → 28.4 dBi 446 MHz → 32.5 dBi
| Cable | 146 MHz | 446 MHz |
|---|---|---|
| RG-58 | 4.6 dB | 8.9 dB |
| RG-8X | 2.7 dB | 5.3 dB |
| LMR-400 | 1.4 dB | 2.7 dB |
50ft RG-8X @ 146 MHz: (50/100) × 2.7 = 1.35 dB
SWR 1.5:1 → 14 dB return loss ← GOOD
SWR 2.0:1 → 9.5 dB ← ACCEPTABLE
SWR 3.0:1 → 6.0 dB ← MARGINAL, tune antenna
Path Loss (dB) = 32.45 + 20log10(f_MHz) + 20log10(dist_km)
146 MHz, 10km → 106.2 dB
Budget = TxPower(dBm) + TxGain - PathLoss - CableLoss - RxNF - SNR_min
50W(47dBm) + 6dBi - 106.2 - 2 - 6 - 10 = -71.2 dB → FAIL
Add 8dBi Yagi → -63.2 dB → marginal, increase height
Negative budget = no link. Redesign before deployment.
VHF 146 MHz: +600 kHz UHF 446 MHz: -5 MHz
CHIRP CSV: Freq,Offset,Duplex,Name,ToneDec,ToneEnc,Mode,Power
Slot 1/2: TDMA doubles capacity vs FM
TGs: 9=Local, 91=Worldwide EN, 3100=USA, 31330=SW, 4000=DMR-MARC
FT8: 8s intervals, 50 Hz BW, LDPC to -20 dB SNR
Freqs: 3.574 (80m), 7.074 (40m), 14.074 (20m) MHz
APRS: 144.39 MHz NA — position beaconing via APRS-IS
Bands: 5.8/3.4 GHz, range 1–20+ km
VLAN: Mgmt(1), Trusted(10), IoT(20), Ham Radio(30), Servers(40), Guest(50)
Rule: Default-deny between VLANs; explicit allows only