freshcrate
Skin:/
Language:English/็ฎ€ไฝ“ไธญๆ–‡
Home > AI Agents > convoke-agents

convoke-agents

Convoke extends BMAD Method AI agents with two types of installable modules: Teams bring new agents for a domain, Skills add new capabilities to existing agents. Install them independently or combine

agenticagentic-aibmad-methodclaude-codeintent-driven-developmentjavascriptmeta-agentpdlcproduct-discovery
Why this rank:Strong adoptionRelease freshnessHealthy release cadence

Description

Convoke extends BMAD Method AI agents with two types of installable modules: Teams bring new agents for a domain, Skills add new capabilities to existing agents. Install them independently or combine them.

README

 โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ•— โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ•— โ–ˆโ–ˆโ–ˆโ•—   โ–ˆโ–ˆโ•—โ–ˆโ–ˆโ•—   โ–ˆโ–ˆโ•— โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ•— โ–ˆโ–ˆโ•—  โ–ˆโ–ˆโ•—โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ•—
 โ–ˆโ–ˆโ•”โ•โ•โ•โ•โ•โ–ˆโ–ˆโ•”โ•โ•โ•โ–ˆโ–ˆโ•—โ–ˆโ–ˆโ–ˆโ–ˆโ•—  โ–ˆโ–ˆโ•‘โ–ˆโ–ˆโ•‘   โ–ˆโ–ˆโ•‘โ–ˆโ–ˆโ•”โ•โ•โ•โ–ˆโ–ˆโ•—โ–ˆโ–ˆโ•‘ โ–ˆโ–ˆโ•”โ•โ–ˆโ–ˆโ•”โ•โ•โ•โ•โ•
 โ–ˆโ–ˆโ•‘     โ–ˆโ–ˆโ•‘   โ–ˆโ–ˆโ•‘โ–ˆโ–ˆโ•”โ–ˆโ–ˆโ•— โ–ˆโ–ˆโ•‘โ–ˆโ–ˆโ•‘   โ–ˆโ–ˆโ•‘โ–ˆโ–ˆโ•‘   โ–ˆโ–ˆโ•‘โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ•”โ• โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ•—
 โ–ˆโ–ˆโ•‘     โ–ˆโ–ˆโ•‘   โ–ˆโ–ˆโ•‘โ–ˆโ–ˆโ•‘โ•šโ–ˆโ–ˆโ•—โ–ˆโ–ˆโ•‘โ•šโ–ˆโ–ˆโ•— โ–ˆโ–ˆโ•”โ•โ–ˆโ–ˆโ•‘   โ–ˆโ–ˆโ•‘โ–ˆโ–ˆโ•”โ•โ–ˆโ–ˆโ•— โ–ˆโ–ˆโ•”โ•โ•โ•
 โ•šโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ•—โ•šโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ•”โ•โ–ˆโ–ˆโ•‘ โ•šโ–ˆโ–ˆโ–ˆโ–ˆโ•‘ โ•šโ–ˆโ–ˆโ–ˆโ–ˆโ•”โ• โ•šโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ•”โ•โ–ˆโ–ˆโ•‘  โ–ˆโ–ˆโ•—โ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ–ˆโ•—
 โ•šโ•โ•โ•โ•โ•โ• โ•šโ•โ•โ•โ•โ•โ• โ•šโ•โ•  โ•šโ•โ•โ•โ•  โ•šโ•โ•โ•โ•   โ•šโ•โ•โ•โ•โ•โ• โ•šโ•โ•  โ•šโ•โ•โ•šโ•โ•โ•โ•โ•โ•โ•
                Agent teams for complex systems

Teams Agents Workflows Skills Version License

Convoke extends AI agents with two types of installable modules: Teams bring new agents for a domain, Skills add new capabilities to existing agents. Install them independently or combine them.

Authoring a Convoke skill? Required reading first: The Convoke Operator Covenant โ€” one axiom and seven Operator Rights every Convoke skill must honor. The Covenant exists because operator-facing UX is an architectural concern in agent systems, not a styling concern; skills that violate it erode operator trust across the skill ecosystem, not just on their own surface.

Team Domain
Vortex Product discovery โ€” from user insight to evidence-based decisions
Gyre Production readiness โ€” from stack detection to gap analysis

What's New in 3.3

  • Convoke Operator Covenant โ€” one axiom ("the operator is the resolver") and seven Operator Rights every Convoke skill must honor. Reproducibility-validated governance contract with a published baseline audit and Compliance Checklist.
  • Opt out of agents you don't need โ€” new excluded_agents: [] field in Vortex and Gyre configs; convoke-update honors exclusions end-to-end. Non-breaking.
  • Update-time changelog surfacing โ€” convoke-update shows "What's New" before confirming the refresh.
  • See the CHANGELOG for the full release details

Vortex โ€” Product Discovery Team

Specialist agents guide you from insight to evidence and back again โ€” a continuous discovery loop, not a one-shot checklist

Agents Workflows

Vortex guides you through seven discovery streams โ€” from understanding your users to interpreting production signals โ€” so you can make evidence-based decisions before, during, and after you build. Each stream builds on the previous one's findings, and when gaps appear, the system routes you back to fill them.

                         7 Streams ยท 7 Agents

  โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”   โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”   โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”   โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”
  โ”‚     Isla    โ”‚โ”€โ”€โ–ถโ”‚     Mila    โ”‚โ”€โ”€โ–ถโ”‚     Liam    โ”‚โ”€โ”€โ–ถโ”‚     Wade    โ”‚
  โ”‚  Empathize  โ”‚   โ”‚ Synthesize  โ”‚   โ”‚ Hypothesize โ”‚   โ”‚ Externalize โ”‚
  โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜   โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜   โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜   โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜
         โ–ฒ                                                       โ”‚
         โ”‚                                                       โ”‚
         โ”‚                                                       โ–ผ
  โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”   โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”   โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”            โ”‚
  โ”‚     Emma    โ”‚โ—€โ”€โ”€โ”‚     Max     โ”‚โ—€โ”€โ”€โ”‚     Noah    โ”‚โ—€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜
  โ”‚Contextualizeโ”‚   โ”‚ Systematize โ”‚   โ”‚  Sensitize  โ”‚
  โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜   โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜   โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜
         โ”‚                 โ”‚                 โ”‚
         โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜
          โ–ถ Start at Emma ยท back to any stream

Each workflow ends with a Compass routing to whichever stream needs attention โ€” you can start or return to any agent.

Agent Stream What they do
Emma ๐ŸŽฏ Contextualize Frame the right problem โ€” personas, product vision, scope
Isla ๐Ÿ” Empathize Understand users โ€” empathy maps, interviews, discovery research
Mila ๐Ÿ”ฌ Synthesize Converge research into clear problem definitions
Liam ๐Ÿ’ก Hypothesize Turn problems into testable hypotheses and experiments
Wade ๐Ÿงช Externalize Test assumptions with MVPs, experiments, and prototypes
Noah ๐Ÿ“ก Sensitize Interpret production signals, user behavior, and engagement patterns
Max ๐Ÿงญ Systematize Capture learnings and decide: pivot, patch, or persevere
22 Vortex Workflows
  • Assumption Mapping
  • Behavior Analysis
  • Contextualize Scope
  • Empathy Map
  • Experiment Design
  • Hypothesis Engineering
  • Lean Experiment
  • Lean Persona
  • Learning Card
  • MVP
  • Pattern Mapping
  • Pivot Patch Persevere
  • Pivot Resynthesis
  • Product Vision
  • Production Monitoring
  • Proof of Concept
  • Proof of Value
  • Research Convergence
  • Signal Interpretation
  • User Discovery
  • User Interview
  • Vortex Navigation

What Agents Produce

Here's a sample of real output from a busy parents meal planning project โ€” each excerpt is from the full 7-agent journey example.

Emma ๐ŸŽฏ Contextualize

Emma frames the right problem. Here's the Job-to-be-Done she produced:

Job-to-be-Done: Eliminate the daily 5:30 PM dinner decision so I can feed my family well without the mental load of planning, shopping, and deciding under time pressure.

Riskiest Assumptions:

  1. Decision fatigue โ€” not cooking skill or ingredient access โ€” is the primary barrier to weeknight dinner success
  2. Parents would trust and act on an externally-provided dinner suggestion rather than needing to choose themselves
  3. "Good enough" nutrition is an acceptable standard โ€” parents aren't seeking perfection, they're seeking relief from guilt

Liam ๐Ÿ’ก Hypothesize

Liam turns problems into testable ideas. Here's one of three hypotheses he produced:

Hypothesis 1: The Pre-Commute Decision Eliminator

We believe that busy parents will act on a single dinner suggestion delivered at 4:00 PM within 3 minutes because the decision burden โ€” not cooking โ€” is their primary barrier, and an earlier intervention catches them before the anxiety spiral begins.

Riskiest Assumption: Parents will trust and act on an automated suggestion without second-guessing. Research shows they want "someone to tell me what to make" โ€” but "someone" may need to be a trusted person, not an algorithm.

Max ๐Ÿงญ Systematize

Max captures what you learned and decides what to do next:

Recommendation: PATCH (iterate on timing, don't pivot direction)

The core hypothesis is validated. The product direction (decision elimination via single suggestion) is correct. The timing mechanism needs refinement โ€” shift from fixed 4:00 PM delivery to adaptive delivery based on each user's observed engagement pattern.

Three Actions:

  1. Implement adaptive timing โ€” shift the push notification to match each user's observed engagement window.
  2. Route to Isla for timing investigation โ€” qualitative research on why users engage at 3:15 PM. Is it anxiety relief, logistical planning, or habit?
  3. Test willingness to pay immediately โ€” the mechanism works, but we have no commercial validation. Run a landing page test with pricing before further product investment.

See the full 7-agent journey example โ†’

For detailed workflow descriptions, see the Agent Guide and individual user guides: Emma | Isla | Mila | Liam | Wade | Noah | Max

Gyre โ€” Production Readiness Team

Specialist agents assess whether your project is ready to ship โ€” detect your stack, model what "ready" means, find what's missing, and refine together

Agents Workflows

Most teams ship to production without knowing if they're actually ready โ€” missing SLOs, incomplete observability, unreviewed deployment strategies. Gyre analyzes your project's filesystem, builds a capabilities model tailored to your specific tech stack, and surfaces what's absent. No generic checklists โ€” every finding is contextual to your project.

                    4 Agents ยท 4 Contracts

  โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”   โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”   โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”   โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”
  โ”‚    Scout    โ”‚โ”€โ”€โ–ถโ”‚    Atlas    โ”‚โ”€โ”€โ–ถโ”‚    Lens     โ”‚โ”€โ”€โ–ถโ”‚    Coach    โ”‚
  โ”‚   Detect    โ”‚   โ”‚   Model     โ”‚   โ”‚   Analyze   โ”‚   โ”‚   Review    โ”‚
  โ”‚   Stack     โ”‚   โ”‚   Capabil.  โ”‚   โ”‚   Gaps      โ”‚   โ”‚   Refine    โ”‚
  โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜   โ””โ”€โ”€โ”€โ”€โ”€โ”€โ–ฒโ”€โ”€โ”€โ”€โ”€โ”€โ”˜   โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜   โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”˜
                           โ”‚                                   โ”‚
                           โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€ GC4 โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜
                                       feedback

Linear pipeline with a feedback loop. Coach amendments feed back to Atlas โ€” the model improves with every review.

Agent Role What they do
Scout ๐Ÿ”Ž Detect Scan filesystem artifacts to classify your tech stack โ€” language, containers, CI/CD, observability, cloud provider
Atlas ๐Ÿ“ Model Generate a capabilities manifest using industry standards (DORA, OpenTelemetry, Google PRR) and web search
Lens ๐Ÿ”ฌ Analyze Compare capabilities against what actually exists โ€” surface absences, compound risks, severity-prioritized findings
Coach ๐Ÿ‹๏ธ Review Walk through findings conversationally โ€” amend the model, capture feedback, track progress over time
7 Gyre Workflows
  • Full Analysis (end-to-end pipeline)
  • Stack Detection
  • Model Generation
  • Gap Analysis
  • Model Review
  • Delta Report
  • Accuracy Validation

What Gyre Produces

Gyre writes structured artifacts to .gyre/ in your project root โ€” safe to commit, containing technology categories only (no file contents, paths, or secrets):

  • Stack Profile (.gyre/stack-profile.yaml) โ€” classified tech stack with detection confidence
  • Capabilities Manifest (.gyre/capabilities.yaml) โ€” 20+ capabilities across observability, deployment, reliability, and security
  • Findings Report (.gyre/findings.yaml) โ€” absence-based findings tagged by severity, confidence, and source
  • Feedback Log (.gyre/feedback.yaml) โ€” team amendments that persist and improve the model over time

Three modes: Crisis (first run, full pipeline), Anticipation (re-analysis with cached model), Regeneration (fresh model rebuild).

Delta tracking: Run Gyre after making improvements to see what you've resolved โ€” findings are tagged [NEW], [CARRIED], or resolved.

User guides: Scout | Atlas | Lens | Coach

Extending Convoke

Team Factory

Create fully-wired, BMAD-compliant teams through a guided workflow โ€” zero post-creation fixes.

/bmad-team-factory

The factory guides you through composition pattern selection (Independent or Sequential), agent scope definition with overlap detection, contract design, and integration wiring. Decision state is persisted as a spec file for resume and express mode. Output passes the same validation as native teams.

Three capabilities:

  • Create Team โ€” build a new team from scratch with full integration
  • Add Agent โ€” extend an existing team with a new agent
  • Add Skill โ€” give an existing agent a new workflow

Portability โ€” Export Skills Anywhere

Take any BMAD skill and export it to a standalone, LLM-agnostic format that works outside Claude Code:

npx convoke-export bmad-brainstorming --output ./exported

The export engine transforms skill workflows into self-contained instruction documents, then generates platform-specific adapter files:

Platform Adapter output (written inside the export target)
Claude {target}/CLAUDE.md commands
GitHub Copilot {target}/.github/copilot-instructions.md
Cursor {target}/.cursor/rules/

Skills are classified by tier: standalone skills export cleanly, light-deps skills include dependency notes, and pipeline skills (multi-step orchestration) are flagged as non-portable.

Four skills support the workflow: bmad-export-skill (export), bmad-validate-exports (validate), bmad-generate-catalog (catalog README), bmad-seed-catalog (full catalog repo).

Enhance โ€” Agent Skills

Skills give existing agents new workflows โ€” installed via menu patching, not agent modification. The first skill adds RICE-scored backlog management to the PM agent:

                    Initiatives Backlog

  โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”   โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”   โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”
  โ”‚  [T] Triage โ”‚   โ”‚  [R] Review โ”‚   โ”‚  [C] Create โ”‚
  โ”‚  Ingest new โ”‚   โ”‚   Rescore   โ”‚   โ”‚  Bootstrap  โ”‚
  โ”‚  findings   โ”‚   โ”‚  existing   โ”‚   โ”‚ from scratchโ”‚
  โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”˜   โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”˜   โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”˜
         โ”‚                 โ”‚                 โ”‚
         โ–ผ                 โ–ผ                 โ–ผ
  โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”
  โ”‚          initiatives-backlog.md                  โ”‚
  โ”‚   RICE-scored ยท Categorized ยท Change-tracked     โ”‚
  โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜

Activate from the PM agent menu or directly: /bmad-enhance-initiatives-backlog

The Enhance Guide documents the complete pattern for building your own skills.

Quick Start

Prerequisites

  • Node.js 18+ or Bun
  • Git
  • Claude Code or Claude.ai

Install

Everything (Vortex + Gyre):

npm install convoke-agents@latest
npx convoke-install

Vortex only:

npm install convoke-agents@latest
npx convoke-install-vortex

Gyre only:

npm install convoke-agents@latest
npx convoke-install-gyre

Something not working? Run npx convoke-doctor or check the FAQ.

Personalize

Open _bmad/bme/_vortex/config.yaml (or _bmad/bme/_gyre/config.yaml) and replace {user} with your name. Agents use this to personalize their interactions.

Activate an Agent

Claude Code (skills)

# Vortex
/bmad-agent-bme-contextualization-expert          # Emma  ๐ŸŽฏ
/bmad-agent-bme-discovery-empathy-expert          # Isla  ๐Ÿ”
/bmad-agent-bme-research-convergence-specialist   # Mila  ๐Ÿ”ฌ
/bmad-agent-bme-hypothesis-engineer               # Liam  ๐Ÿ’ก
/bmad-agent-bme-lean-experiments-specialist        # Wade  ๐Ÿงช
/bmad-agent-bme-production-intelligence-specialist # Noah  ๐Ÿ“ก
/bmad-agent-bme-learning-decision-expert           # Max   ๐Ÿงญ

# Gyre
/bmad-agent-bme-stack-detective                    # Scout ๐Ÿ”Ž
/bmad-agent-bme-model-curator                      # Atlas ๐Ÿ“
/bmad-agent-bme-readiness-analyst                  # Lens  ๐Ÿ”ฌ
/bmad-agent-bme-review-coach                       # Coach ๐Ÿ‹๏ธ

Claude Code (terminal) / Other AI assistants

# Vortex
cat _bmad/bme/_vortex/agents/contextualization-expert.md    # Emma  ๐ŸŽฏ
# ... (see user guides for full list)

# Gyre
cat _bmad/bme/_gyre/agents/stack-detective.md               # Scout ๐Ÿ”Ž
cat _bmad/bme/_gyre/agents/model-curator.md                 # Atlas ๐Ÿ“
cat _bmad/bme/_gyre/agents/readiness-analyst.md             # Lens  ๐Ÿ”ฌ
cat _bmad/bme/_gyre/agents/review-coach.md                  # Coach ๐Ÿ‹๏ธ

Claude.ai โ€” Open any agent file and paste its contents into your conversation.

How activation works: Each agent is a markdown file containing a full persona, menu system, and workflow instructions. When Claude reads the file, it adopts that agent's expertise and presents you with an interactive menu. Pick a workflow from the menu and follow the guided steps.

Your First 15 Minutes

  1. Personalize โ€” edit the config.yaml for your chosen team and replace {user} with your name
  2. Pick a starting point:
    • Vortex: Activate Emma โ†’ select Lean Persona from the menu โ†’ follow the guided steps
    • Gyre: Activate Scout โ†’ select Full Analysis from the menu โ†’ walk through the pipeline
  3. Find your artifact โ€” outputs are saved in _bmad-output/vortex-artifacts/ or .gyre/
  4. Follow the Compass โ€” each workflow ends with a routing suggestion for which agent to use next

What Gets Installed

your-project/
โ”œโ”€โ”€ _bmad/bme/
โ”‚   โ”œโ”€โ”€ _vortex/              # Team: Product Discovery
โ”‚   โ”‚   โ”œโ”€โ”€ agents/           # 7 agent definition files
โ”‚   โ”‚   โ”œโ”€โ”€ workflows/        # 22 workflows
โ”‚   โ”‚   โ”œโ”€โ”€ contracts/        # Handoff contracts (HC1-HC5 artifact, HC6-HC10 routing)
โ”‚   โ”‚   โ”œโ”€โ”€ guides/           # User guides (all 7 agents)
โ”‚   โ”‚   โ””โ”€โ”€ config.yaml       # Configuration
โ”‚   โ”œโ”€โ”€ _gyre/                # Team: Production Readiness
โ”‚   โ”‚   โ”œโ”€โ”€ agents/           # 4 agent definition files
โ”‚   โ”‚   โ”œโ”€โ”€ workflows/        # 7 workflows
โ”‚   โ”‚   โ”œโ”€โ”€ contracts/        # Artifact contract schemas (GC1-GC4)
โ”‚   โ”‚   โ”œโ”€โ”€ guides/           # User guides (all 4 agents)
โ”‚   โ”‚   โ””โ”€โ”€ config.yaml       # Configuration
โ”‚   โ”œโ”€โ”€ _enhance/             # Skill: Agent Capability Upgrades
โ”‚   โ”‚   โ”œโ”€โ”€ workflows/        # Skill workflows (initiatives-backlog)
โ”‚   โ”‚   โ”œโ”€โ”€ extensions/       # Agent menu patch descriptors
โ”‚   โ”‚   โ”œโ”€โ”€ guides/           # Module author guide
โ”‚   โ”‚   โ””โ”€โ”€ config.yaml       # Configuration
โ”‚   โ”œโ”€โ”€ _portability/         # Skill: Export skills to other platforms
โ”‚   โ”‚   โ””โ”€โ”€ skills/           # Export, validate, catalog, seed workflows
โ”‚   โ”œโ”€โ”€ _team-factory/        # Skill: Create new BMAD-compliant teams
โ”‚   โ”‚   โ””โ”€โ”€ lib/              # Factory generators and validators
โ”‚   โ””โ”€โ”€ _artifacts/           # Skill: Artifact governance & portfolio
โ”‚       โ””โ”€โ”€ workflows/        # Migrate artifacts, portfolio status
โ””โ”€โ”€ _bmad-output/
    โ”œโ”€โ”€ vortex-artifacts/     # Vortex generated artifacts
    โ””โ”€โ”€ gyre-artifacts/       # Gyre generated artifacts

How It Fits with BMAD Core

Convoke handles discovery, validation, and readiness. BMAD Core handles implementation.

Convoke Modules                            BMAD Core
โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”          โ”Œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”
โ”‚ Teams                        โ”‚          โ”‚                      โ”‚
โ”‚   Vortex (Product Discovery) โ”‚ โ”€โ”€โ”€โ”€โ”€โ”€>  โ”‚ PM โ†’ Architect โ†’ Dev โ”‚
โ”‚   Gyre (Prod. Readiness)     โ”‚ <โ”€โ”€โ”€โ”€โ”€โ”€  โ”‚ "Let's build it"     โ”‚
โ”‚                              โ”‚          โ”‚                      โ”‚
โ”‚ Skills                       โ”‚  signals โ”‚                      โ”‚
โ”‚   Enhance (Agent Upgrades)   โ”‚ โ”€โ”€โ”€โ”€โ”€โ”€>  โ”‚                      โ”‚
โ”‚                              โ”‚          โ”‚                      โ”‚
โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜          โ””โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”˜

Teams and Skills are peer module types โ€” both installable, both independent. Convoke works standalone or as an extension โ€” no BMAD Method installation required.

Updating

npm install convoke-agents@latest         # Get the latest package
npx convoke-version                       # Check current version
npx convoke-update --dry-run              # Preview changes
npx convoke-update                        # Apply update (auto-backup)
npx convoke-doctor                        # Diagnose issues

Your data in _bmad-output/ and .gyre/ is never touched. Automatic backups are created before every update.

Important: npm install convoke-agents (without @latest) won't cross major version boundaries. If you're on v2.x, you must use npm install convoke-agents@latest to get v3.x.

See UPDATE-GUIDE.md for migration paths and troubleshooting.

Documentation

Document Description
Agent Guide Detailed agent descriptions, workflows, and positioning
BMAD Compatibility Compatibility matrix with BMAD Method versions
Testing Automated test suite, CI pipeline, and agent test results
Development Guide Architecture, building agents, and contributing
FAQ Common questions about the framework
CHANGELOG Complete version history
UPDATE-GUIDE Migration paths and update troubleshooting

Roadmap

  • v1.x โ€” Vortex foundation: 7 agents, 22 workflows, update system, CI/CD
  • v2.0 โ€” Product renamed to Convoke. CLI commands: convoke-*. Package: convoke-agents
  • v2.x โ€” Enhance module (Skills architecture, RICE initiatives-backlog), Gyre team (production readiness, 4 agents), Team Factory
  • v3.x โ€” Artifact governance, portfolio intelligence, portability system (export to Claude/Copilot/Cursor), 1,123 tests
  • Next โ€” Forge (domain knowledge extraction for enterprise brownfield engagements)
  • Future โ€” Additional teams, Forge-Gyre integration, cross-team workflows

Contributing

We welcome contributions in these areas:

  • Agents โ€” New domain-specialized agents, workflow improvements
  • Teams โ€” New team modules via the Team Factory
  • Testing โ€” Edge cases, performance testing
  • Documentation โ€” Tutorials, translations, video walkthroughs

Have feedback? Found a quality issue, want a missing capability, or have a general comment about an agent? Open an issue and select the Agent/Workflow Feedback template.

See the Development Guide for architecture details and agent development patterns.

License

MIT License โ€” see LICENSE

Acknowledgments

Agents

Every agent below contributed to the design, implementation, testing, or documentation of this release.

Core

  • BMad Master ๐Ÿง™ โ€” Orchestration and agent coordination

BMM โ€” BMAD Method Module

  • Mary ๐Ÿ“Š Analyst โ€” Requirements analysis and product briefs
  • Winston ๐Ÿ—๏ธ Architect โ€” Architecture decisions and technical design
  • Amelia ๐Ÿ’ป Dev โ€” Implementation and code delivery
  • John ๐Ÿ“‹ PM โ€” Product management and PRD creation
  • Barry ๐Ÿš€ Quick Flow โ€” Rapid solo development
  • Quinn ๐Ÿงช QA โ€” Quality assurance and test validation
  • Bob ๐Ÿƒ SM โ€” Sprint management and retrospectives
  • Paige ๐Ÿ“š Tech Writer โ€” Documentation and editorial review
  • Sally ๐ŸŽจ UX Designer โ€” User experience design

BME โ€” Vortex Pattern Agents

  • Emma ๐ŸŽฏ Contextualize โ€” Problem framing and product vision
  • Isla ๐Ÿ” Empathize โ€” User research and empathy mapping
  • Mila ๐Ÿ”ฌ Synthesize โ€” Research convergence and problem definition
  • Liam ๐Ÿ’ก Hypothesize โ€” Hypothesis engineering and assumption mapping
  • Wade ๐Ÿงช Externalize โ€” Lean experiments and MVP testing
  • Noah ๐Ÿ“ก Sensitize โ€” Production intelligence and signal interpretation
  • Max ๐Ÿงญ Systematize โ€” Learning capture and pivot/persevere decisions

BME โ€” Gyre Pattern Agents

  • Scout ๐Ÿ”Ž Stack Detective โ€” Technology stack detection and classification
  • Atlas ๐Ÿ“ Model Curator โ€” Contextual capabilities manifest generation
  • Lens ๐Ÿ”ฌ Readiness Analyst โ€” Absence detection and cross-domain correlation
  • Coach ๐Ÿ‹๏ธ Review Coach โ€” Guided review, amendment, and feedback capture

CIS โ€” Creative & Innovation Strategies

  • Carson ๐Ÿง  Brainstorming Coach โ€” Creative ideation facilitation
  • Dr. Quinn ๐Ÿ”ฌ Problem Solver โ€” Systematic problem-solving
  • Maya ๐ŸŽจ Design Thinking Coach โ€” Human-centered design
  • Victor โšก Innovation Strategist โ€” Business model innovation
  • Caravaggio ๐ŸŽจ Presentation Master โ€” Visual communication
  • Sophia ๐Ÿ“– Storyteller โ€” Narrative and storytelling

BMB โ€” BMAD Builder Module

  • Bond ๐Ÿค– Agent Builder โ€” Agent creation and configuration
  • Morgan ๐Ÿ—๏ธ Module Builder โ€” Module architecture
  • Wendy ๐Ÿ”„ Workflow Builder โ€” Workflow design and validation

TEA โ€” Test Engineering & Architecture

  • Murat ๐Ÿงช Test Architect โ€” Test framework, CI pipeline, and quality gates

Discover what has to be discovered. Ship what's ready to ship.

Get Started | Agents | Docs | Roadmap

Release History

VersionChangesUrgencyDate
v3.3.0Most agent frameworks measure themselves by capability density: more agents, more tools, more integrations. Convoke 3.3.0 commits, in code, to a different axis โ€” operator experience as a first-class architectural concern. The new Convoke Operator Covenant is an enforceable contract: one axiom ("the operator is the resolver") and seven operator rights every Convoke skill must honor. Not a tagline โ€” dogfooded against our own skills, scored against a public Compliance Checklist, and reproducibilityHigh4/19/2026
v3.2.0<p data-line="2" class="code-line" dir="auto" style="margin-top: 0px; margin-bottom: 16px; position: relative; color: rgb(204, 204, 204); font-family: -apple-system, &quot;system-ui&quot;, &quot;Segoe WPC&quot;, &quot;Segoe UI&quot;, system-ui, Ubuntu, &quot;Droid Sans&quot;, sans-serif; font-size: 14px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: start; text-indent: 0px; text-transform: none; wiHigh4/11/2026
v3.1.0<h1 style="color: rgb(204, 204, 204); font-family: -apple-system, &quot;system-ui&quot;, &quot;Segoe UI&quot;, Roboto, sans-serif; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; letter-spacing: normal; orphans: 2; text-align: start; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; white-space: normal; background-color: rgb(24, 24, 24); text-decoration-thickness: initial; text-decoration-style: initial; text-decoHigh4/6/2026
v3.0.0<p style="white-space: pre-wrap; margin-top: 0.1em; margin-bottom: 0.2em; color: rgb(204, 204, 204); font-family: -apple-system, &quot;system-ui&quot;, &quot;Segoe UI&quot;, Roboto, sans-serif; font-size: 13px; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: start; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; background-color: rgb(24, 24, 24); Medium3/25/2026
v2.4.0Convoke now has two types of installable modules: Teams bring new agents for a domain, Skills add new capabilities to existing agents. Both are top-level, peer module types. Initiatives Backlog Skill (PM Agent) The first Enhance skill adds RICE-scored backlog management to the PM agent with 3 modes: Triage โ€” Ingest review findings, extract actionable items, propose RICE scores with two-gate validation, append to existing backlog Review โ€” Walk through existing items one at a time, rescoreLow3/15/2026
v2.3.1## Summary - **Migration chaining bug fix** (v2.3.1) โ€” critical fix for multi-version upgrades - **Skills migration** (v2.2.0) โ€” agent activation moved from commands to skills (BMAD v6.1.0) - **Documentation fixes** (v2.3.1) โ€” corrected package names, npx flags, archived stale docs ## What's Changed ### ๐Ÿ”ง Bug Fix: Migration Chaining (v2.3.1) `getMigrationsFor()` only matched the first migration hop โ€” users jumping multiple versions (e.g., 1.0.x โ†’ 2.x) silently skipped all intermediaLow3/15/2026
v2.0.0Migration: npm install convoke-agents โ€” your _bmad/ and _bmad-output/ data is fully preserved. Breaking Changes: CLI commands renamed (bmad-* โ†’ convoke-*). See CHANGELOG.md for details. The old bmad-enhanced package is deprecated on npm with a migration notice.Low3/7/2026
v1.7.0## Highlights - **Fixed npm package bloat** โ€” 1618 files / 9.7 MB โ†’ 242 files / 1.1 MB. Internal backups, migration logs, and non-Vortex BMAD modules were leaking into the published package. - **18 production-quality workflow steps** โ€” Wade's lean-experiment, proof-of-concept, and proof-of-value workflows are now fully authored. All 22 workflows are complete. - **Config merger smart-merge** โ€” `bmad-update` now preserves user-added agents and workflows instead of overwriting them. - **READMLow3/3/2026
v1.6.0**The Innovation Vortex is now complete.** BMAD-Enhanced v1.6.0 delivers the full 7-stream [Innovation Vortex](https://unfix.com/innovation-vortex) pattern (from the [unFIX model](https://unfix.com/) by [Jurgen Appelo](https://jurgenappelo.com/)) โ€” three new agents, nine new workflows, ten handoff contracts, and cross-agent routing that connects every stream into a non-linear product discovery system. ## What's New ### 3 New Agents | Agent | Stream | Role | |-------|--------|------| |Low2/26/2026
v1.5.0## Wave 2 โ€” Four Vortex Streams Now Live BMAD-Enhanced now covers 4 of 7 Innovation Vortex streams with two new agents joining Emma and Wade. ### New Agents - **Isla ๐Ÿ” โ€” Discovery & Empathy Expert** (Empathize stream) - Empathy Map โ€” build deep user understanding through structured empathy mapping - User Interview โ€” design interview scripts, conduct interviews, capture findings - User Discovery โ€” plan and execute discovery research with synthesis - **Max ๐Ÿงญ โ€” Learning & DecLow2/20/2026
v1.4.1<h3 style="color: rgb(204, 204, 204); font-family: -apple-system, &quot;system-ui&quot;, &quot;Segoe UI&quot;, Roboto, sans-serif; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; letter-spacing: normal; orphans: 2; text-align: start; text-indent: 0px; text-transform: none; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; white-space: normal; background-color: rgb(24, 24, 24); text-decoration-thickness: initial; text-decoration-style: initial; text-decoLow2/19/2026
v1.4.0## What's Changed v1.4.0 is an internal architecture refactor of the update/migration system. No user-facing breaking changes. ### Architecture Improvements - **Single source of truth for version** - Version read from `package.json` at runtime, never hardcoded - **Migrations = deltas + shared refresh** - Migration files contain only version-specific logic; shared `refreshInstallation()` handles all file copying - **Append-only registry** - No `toVersion` field, target is always current Low2/18/2026
v1.3.8## What's New Complete Vortex Framework with update/migration system, all 7 workflows, and standalone installation support. ### Highlights - **7 workflows fully implemented** (56 files) - Emma: lean-persona, product-vision, contextualize-scope - Wade: mvp, lean-experiment, proof-of-concept, proof-of-value - **Update/migration system** - `npx bmad-update`, `npx bmad-version`, `npx bmad-migrate` - **Automatic backups** before every migration with rollback on failure - **StandalonLow2/18/2026
v1.0.0-alpha# BMAD-Enhanced v1.0.0-alpha **First public release of BMAD-Enhanced!** ๐ŸŽ‰ Domain-specialized agents for BMAD Method - expert agents for design, quality, and standards workflows. ## โœจ What's Included ### Agents (2) - **Emma (empathy-mapper)** - Empathy Mapping Specialist - **Wade (wireframe-designer)** - Wireframe Design Expert ### Features - โœ… One-command installation (`npm install bmad-enhanced && npm run install:agents`) - โœ… BMAD Method prerequisite checking - โœ… 100% test Low2/15/2026

Dependencies & License Audit

Loading dependencies...

Similar Packages

pickle-rick-claude๐Ÿฅ’ Pickle Rick for Claude Code โ€” autonomous PRD-driven coding loops + relentless code review. Ralph Loop toolkit.v1.103.0
maestro-orchestrateMulti-agent orchestration platform for Gemini CLI and Claude Code โ€” 22 specialists, parallel subagents, persistent sessions, and built-in code review, debugging, security, SEO, accessibility, and compv1.6.4
20xSelf-improving Agent orchestrator for all knowledge workv0.0.94
cc-caffeineโ˜• Keep your laptop awake during commutes and breaks, ensuring uninterrupted coding sessions with cc-caffeine for smarter work and more freedom.main@2026-06-05
be-my-butlerOrchestrate multiple agents to execute Claude Code workflows with cross-model verification for reliable AI code automation.main@2026-06-05

More in AI Agents

@blockrun/franklinFranklin โ€” The AI agent with a wallet. Spends USDC autonomously to get real work done. Pay per action, no subscriptions.
hermes-agentThe agent that grows with you
awesome-copilotCommunity-contributed instructions, agents, skills, and configurations to help you make the most of GitHub Copilot.
e2bE2B SDK that give agents cloud environments