By haabe
Enforce a theory-grounded product discovery and delivery discipline for AI agents, guiding projects from purpose to market through evidence-gated diamond stages, opportunity solution trees, user research, threat modeling, and continuous health monitoring.
Systematically challenge current assumptions before major decisions. Counters confirmation bias, groupthink, and overconfidence.
Use to evaluate the current state of a diamond. Checks theory gates, confidence levels, and recommends next action.
Progress a diamond from one phase to the next. Runs all required theory gate checks, validates evidence, and at Deliver->Complete runs the executable Definition of Done checklist.
Assess delivery health metrics. For software: DORA + APEX. For content/AI/service products: product-type-appropriate metrics.
Run eval scenarios to benchmark Mycelium effectiveness. Execute tasks using reflexion loop, validate against success criteria, record metrics.
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Your AI agent should think before it codes.
AI has made building cheap. It hasn't made deciding cheap. Agents will jump from an idea to a pull request without asking why, who for, or whether anyone needs it. Other tools accelerate delivery — Mycelium makes the agent earn the right to start.
# Recommended (post-v0.20.0): install as a Claude Code plugin
/plugin marketplace add haabe/mycelium
/plugin install mycelium@haabe-mycelium
/mycelium:start # one command: setup + 10-minute discovery
Plugin install is brownfield-safe: no project-root files are modified. Skills are namespaced as /mycelium:<name>. See docs/get-started.md for details.
On the namespace prefix. Anthropic's plugin convention requires /<plugin>:<skill>, so every Mycelium skill is /mycelium:foo. Two ergonomics that take the typing tax down:
/myc<Tab> in Claude Code and it expands to /mycelium:. Then a few letters of the skill name + <Tab> finishes it. /mycelium:diamond-assess is six keystrokes./... form is faster once you know the name; prose is fine when you don't.Legacy install (pre-v0.20.0, still supported during transition):
npx degit haabe/mycelium my-project && cd my-project
# Start Claude Code, then:
/interview
Build to learn, then build to earn (Patton). 30+ established frameworks, connected by theory gates so critical steps cannot be skipped. The agent does not progress until the evidence says it should. Discovery to market feedback at six scales (Purpose → Strategy → Opportunity → Solution → Delivery → Market), the same four-phase diamond at every scale. Configuration files plus orchestrated prompts — not a software library.
Builders — solo developers or small teams using AI agents to build products. If you can't afford to burn runway on the wrong thing, Mycelium helps you find the right thing before you build it.
Works for software, online courses, AI tools, and services. One command to start. The agent guides you from there.
Mycelium is for work where deciding what to build is the hard part. Some use cases are better served elsewhere — saying so up front saves frustration:
Time-constrained projects ARE supported as of 2026-04-30: /interview Phase 0 picks <8h inline discovery, 8-48h sprint mode, or 48h+ full interview. The path is selected by your answer to "How much time do you have?"
Not 45 skills dumped on you at once. Three modes that show up at the right time:
| When | Experience | Example |
|---|---|---|
| During a phase | Mentor | "Have you considered who your real user is? Here's what the research says about purpose statements." |
| At boundaries | Guardrail | "You're about to skip the bias check. The evidence gate requires this before progressing." |
| At transitions | Checklist | "Before moving forward: evidence ✓, bias check ✗, corrections ✓" |
A small project sees fewer gates and lighter guidance. A complex product gets the full treatment.
Mycelium has been dogfooded on three small projects AND tested by one outside user under realistic time pressure. Each session taught the framework something different — and most of what they taught is in the version you're looking at right now.
npx claudepluginhub haabe/mycelium --plugin myceliumAgent-first PM toolkit with 9 specialist agents and 18 skills for solo developers and small teams
Strategic thinking partner for product decisions. Works through problems conversationally, challenges assumptions, helps you ship faster. Grounded in frameworks from Marty Cagan, Teresa Torres, Elena Verna, Brian Balfour, Chip Huyen, Ryan Singer, Hamel Husain, and more. Complete eval chain from first 20 test cases through error analysis, LLM judges, and RAG evaluation. Plus backlog automation with Linear/GitHub integration.
📋 Product Manager — Investigative Product Strategist + Market-Savvy PM
Opportunity Solution Tree framework for pre-PMF startups to guide product discovery, problem validation, and solution prioritization
Advanced PM skills: AI Product Canvas, Multi-Source Signal Synthesiser, Experiment Designer, Design Handoff Brief. For senior PMs working on complex or AI-powered products.
This skill should be used when users need to generate ideas, explore creative solutions, or systematically brainstorm approaches to problems. Use when users request help with ideation, content planning, product features, marketing campaigns, strategic planning, creative writing, or any task requiring structured idea generation. The skill provides 30+ research-validated prompt patterns across 14 categories with exact templates, success metrics, and domain-specific applications.