agency-orchestrator

agency-orchestrator is a TypeScript multi-agent orchestration engine that runs YAML-defined DAG workflows using prebuilt role prompts (186 roles). It can execute steps in parallel based on dependencies, pass outputs between steps via variables, and supports execution via multiple LLM providers/CLIs (including subscription-based “no API key” flows) as well as an MCP server mode for tool-calling from AI coding clients.

Evaluated Mar 30, 2026 (0d ago)
Homepage ↗ Repo ↗ Ai Ml ai-agents agent-orchestration yaml-workflow dag mcp typescript multi-agent cli
⚙ Agent Friendliness
65
/ 100
Can an agent use this?
🔒 Security
48
/ 100
Is it safe for agents?
⚡ Reliability
40
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
70
Documentation
80
Error Messages
0
Auth Simplicity
65
Rate Limits
30

🔒 Security

TLS Enforcement
70
Auth Strength
45
Scope Granularity
20
Dep. Hygiene
55
Secret Handling
50

Security posture is largely dependent on underlying provider CLIs/APIs (e.g., how they handle auth tokens) and on whether prompts/inputs (including user data) are sent to third-party services. The provided materials mention retries and caching/resume, but do not document secure secret storage, logging redaction, or least-privilege scopes for model access. TLS and secret-handling specifics are not explicitly stated.

⚡ Reliability

Uptime/SLA
0
Version Stability
45
Breaking Changes
40
Error Recovery
75
AF Security Reliability

Best When

You want to compose multi-agent workflows quickly (YAML DAGs), reuse prebuilt role prompts, and run within the capabilities of supported providers/CLIs or via MCP from IDE/coding agents.

Avoid When

You need strict, auditable enterprise security controls, fine-grained authorization scopes, or a clearly documented stable public HTTP API for integration.

Use Cases

  • Multi-step product/tech/design reviews with role specialization
  • PRD-to-requirements and requirements-to-plan generation
  • Code review and security/performance audit workflows
  • Content pipelines (e.g., marketing posts, long-form articles)
  • Strategy/ops templates (incident postmortems, weekly reports, hiring pipelines)
  • Rapid prototyping of agent workflows using YAML and built-in templates
  • Interactive iteration using resume/re-run from prior step outputs

Not For

  • Applications requiring a guaranteed, single-vendor hosted API contract
  • Production systems needing strict compliance guarantees (no documented residency/compliance controls in provided materials)
  • Environments where sending prompts to third-party model providers is unacceptable
  • Use cases needing first-class REST/GraphQL web APIs for external service integration

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
Yes
Webhooks
No

Authentication

Methods: Provider-specific subscription/CLI authentication (e.g., claude-code, gemini-cli, copilot-cli, codex-cli, openclaw-cli) API-key based providers via environment variables (OPENAI_API_KEY, ANTHROPIC_API_KEY, DEEPSEEK_API_KEY, and OpenAI-compatible base_url/api_key) Local model via Ollama (typically local runtime authorization/none)
OAuth: No Scopes: No

No first-party OAuth is described; authentication is delegated to the underlying provider CLIs/APIs. The README claims “no API key needed” for certain provider CLIs when the user has an existing subscription.

Pricing

Free tier: Yes
Requires CC: No

README emphasizes zero additional API-key cost for supported subscription flows, but does not specify official hosted pricing for the orchestrator itself.

Agent Metadata

Pagination
none
Idempotent
True
Retry Guidance
Documented

Known Gotchas

  • Behavior depends heavily on the selected provider/CLI; provider-specific rate limits and auth errors may surface differently.
  • DAG parallelism may expose race conditions if steps have implicit dependencies not captured via depends_on/variables.
  • Resuming relies on prior output files/metadata.json; using incorrect directories may lead to missing variables or partial reruns.
  • When using MCP from IDE tools, ensure the command/args match the installed package version and runtime environment.

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for agency-orchestrator.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-03-30.

6533
Packages Evaluated
19870
Need Evaluation
586
Need Re-evaluation
Community Powered