5,501,259 AI resources indexedWorking toward the world’s largest AI resource index.

PDF

ideate - Discovery Brief Methodology

Internal methodology for the ideation discovery stage. Produces a concise Discovery Brief for downstream research, architecture, PRD, and planning.

Installable skillApache-2.0By GobbyAIUpdated Oct 28, 2026
INTENDED FORClaude CodeCodexOpenClawCursorInstall in a listed agent.

instructions.md

---
name: ideate-8f66c0059de3
description: Internal methodology for the ideation discovery stage. Produces a concise Discovery Brief for downstream research, architecture, PRD, and planning.
version: "1.0.0"
category: methodology
internal: true
triggers: ideation, discovery brief, problem framing
metadata:
  gobby:
    audience: agent
    depth: 0
---

# ideate - Discovery Brief Methodology

Use this skill when acting as the `analyst` discovery-stage agent.

## Purpose

Frame the work before evidence gathering. Convert the task description, user
context, existing labels, and any visible constraints into a short brief that
downstream stages can rely on.

## Method

1. Read the assigned task and existing discovery marker blocks.
2. Identify the real problem, not just the requested change.
3. Separate known constraints from assumptions.
4. State the most likely solution hypotheses without committing architecture.
5. Capture open questions only when answering them would change the work.

## Output

Write only this section shape:

```markdown
## Discovery Brief

### Problem
- <what needs to be solved>

### Constraints
- <known constraint>

### Hypotheses
- <testable direction>

### Open Questions
- <question or "None">
```

Keep the brief practical and evidence-seeking. Do not write implementation
tasks, PRD scope, architecture decisions, or planning manifests.

Discuss this resource

Platform guidance

Imported from the listed AgentSkill.sh category. Complete supplied package files and notices are retained. Required external tools and credentials are not installed. No bundled code was executed during import. Individual agent task behavior has not been tested.