grimoire-innovate

name	grimoire-innovate
description	Run the Grimoire innovation pipeline from idea to prototype. Use when: new idea, innovation, brainstorm to code, incubator, R&D cycle, experiment, explore alternative, what-if scenario, quantum branch, parallel timelines, prototype, seed idea. Combines incubator lifecycle, R&D engine, and quantum branching.

Grimoire Innovate

Full innovation pipeline: submit ideas, evaluate viability, run R&D cycles, explore in isolated branches.

When to Use

• New idea that needs structured evaluation
• Running an innovation sprint
• When the user says "what if...", "let's experiment", "new idea", or "R&D"
• Evaluating competing approaches
• Exploring architectural alternatives safely

When NOT to Use

• For a one-off divergent ideation session without R&D follow-through → grimoire-brainstorming.
• To turn a chosen idea into a step-by-step implementation plan → grimoire-writing-plans.
• This skill is the full pipeline (idea → incubator → R&D → prototype); skip it for narrow lookups.

Procedure

Step 1 — Check Incubator Status

python3 framework/tools/incubator.py --project-root . status

See what ideas are already in the pipeline and their lifecycle stages:

• SEED → INCUBATING → VIABLE → HATCHED
• DORMANT (parked), DEAD (abandoned)

Step 2 — Submit or Wake Idea

New idea:

python3 framework/tools/incubator.py --project-root . submit \
  --title "<idea title>" \
  --description "<detailed description>"

Wake a dormant idea:

python3 framework/tools/incubator.py --project-root . wake --id <IDEA-NNN>

Step 3 — Viability Check

python3 framework/tools/incubator.py --project-root . viable --id <IDEA-NNN>

Evaluates:

• Alignment with project objectives
• Technical feasibility (dependencies satisfied?)
• Duplication with existing features
• Sponsor validation (at least one agent approved)

Step 4 — R&D Cycle (for viable ideas)

Quick evaluation:

python3 framework/tools/r-and-d.py --project-root . cycle --quick

Runs harvest + evaluate phases only.

Full cycle (7 phases):

python3 framework/tools/r-and-d.py --project-root . cycle

Phases: HARVEST → EVALUATE → CHALLENGE → SIMULATE → IMPLEMENT → SELECT → CONVERGE

Intensive training (multiple cycles):

python3 framework/tools/r-and-d.py --project-root . train --epochs 3

Uses reinforcement learning to optimize innovation parameters across cycles.

Step 5 — Quantum Branch (for exploration)

If the idea needs isolated experimentation:

Fork a parallel timeline:

python3 framework/tools/quantum-branch.py --project-root . fork --name "<experiment-name>"

List active branches:

python3 framework/tools/quantum-branch.py --project-root . list

Compare branches:

python3 framework/tools/quantum-branch.py --project-root . compare --branches main,<experiment-name>

Merge successful experiment:

python3 framework/tools/quantum-branch.py --project-root . merge --source <experiment-name>

Prune failed experiment:

python3 framework/tools/quantum-branch.py --project-root . prune --branch <experiment-name>

Step 6 — Innovation Dashboard

python3 framework/tools/r-and-d.py --project-root . dashboard

Step 7 — Report

## 💡 Innovation Pipeline Report

### Incubator Status
| Stage      | Count | Key Ideas              |
|------------|-------|------------------------|
| SEED       | N     | ...                    |
| INCUBATING | N     | ...                    |
| VIABLE     | N     | ...                    |
| DORMANT    | N     | ...                    |
| HATCHED    | N     | (completed)            |

### R&D Cycle Results
- Cycles run: N
- Ideas evaluated: N
- Surviving ideas: N
- Top innovation: <title> (fitness: X)

### Active Experiments (Quantum Branches)
| Branch | Purpose | Age | Status |
|--------|---------|-----|--------|
| ...    | ...     | Nd  | active/ready-to-merge |

### Recommendations
1. [Ideas ready for implementation]
2. [Branches ready to merge]
3. [Dormant ideas worth revisiting]

Quick Mode

Fast idea-to-evaluation without branches:

python3 framework/tools/incubator.py --project-root . submit --title "<idea>" --description "<desc>"
python3 framework/tools/r-and-d.py --project-root . cycle --quick

Notes

• Quantum branches are filesystem-level snapshots, not git branches
• R&D cycles use reinforcement learning — results improve over multiple runs
• Incubator prune will clean ideas that have been SEED for too long without evaluation
• All tools support --json for programmatic output