Skip to content
Back to skills

Blueprint Protocol

ASecurity

Planning workflow - OBSERVE → ORIENT → PLAN → APPROVE → EXECUTE → DOCUMENT

  • 5 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added June 6, 2026
testingtypescriptgodocumentation

Works with

  • cli

Security analysis

A100/100

Scanned June 6, 2026

npx -y skills add AxelMrak/ai --skill blueprint-protocol --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Blueprint Protocol?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Blueprint Protocol
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/axelmrak-blueprint-protocol/badge)](https://www.skillsdirectory.com/skills/axelmrak-blueprint-protocol)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: blueprint-protocol
description: Planning workflow - OBSERVE → ORIENT → PLAN → APPROVE → EXECUTE → DOCUMENT
---

# Blueprint Protocol

> The non-negotiable workflow for non-trivial tasks.

## Cycle

```
OBSERVE → ORIENT → PLAN → APPROVE → EXECUTE → DOCUMENT
```

### 1. OBSERVE
Read `.ai/` files to understand current state:
- `.ai/CONTEXT.md` - Project fundamentals
- `.ai/checkpoints/LATEST.md` - Current focus
- `.ai/plans/` - Active plans
- `.ai/TO-DO.md` - Pending work

### 2. ORIENT
Analyze request against architecture:
- Identify gaps
- Check constraints
- Verify alignment with existing patterns

### 3. PLAN
Propose approach with:
- **Options**: A) Simple vs B) Scalable
- **Recommendation**: Explicit choice with justification
- **Steps**: Atomic, actionable items
- **Risks**: What could go wrong

### 4. APPROVE
**CRITICAL:** STOP and wait for explicit approval

Ask: "¿Le mando mecha?" / "Shall I execute?"

Never execute without:
- "dale" / "go ahead" / "execute"
- Explicit user approval

### 5. EXECUTE
Implement only after approval:
- Follow plan step-by-step
- Document deviations
- Update progress

### 6. DOCUMENT
Update working memory:
- Plan status
- `.ai/checkpoints/LATEST.md`
- `.ai/TO-DO.md`

## Anti-Patterns

❌ Skip approval and execute immediately
❌ Plan without reading existing context
❌ Execute without updating documentation
❌ Assume approval from silence

## Examples

**ATHENA (Planning):**
```markdown
## Analysis
Current auth uses JWT but lacks refresh tokens

## Options
**A) Simple:** Add basic refresh token rotation
**B) Scalable:** Implement OAuth2 with PKCE

## Recommendation
Option A - meets current needs, OAuth overkill for v1

## Plan
1. [ ] Add refresh token to JWT payload
2. [ ] Create /auth/refresh endpoint
3. [ ] Update client to auto-refresh

---
¿Aprobás? Lo paso a APOLLO para implementación.
```

**APOLLO (Execution):**
```markdown
## Executing Plan: auth-refresh
Step 1/3: Add refresh token to JWT

### Implementation
- Modified jwt.ts to include refresh_token claim
- Updated types to include refresh payload

### Verification
- [x] TypeScript compiles
- [x] Tests pass

---
¿Continúo con Step 2?
```

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.