prompt
Orchestrator skill for the `prompt` skillset. Dispatches to member skills in a safe, predictable order. Separates intent formation from execution to protect humans from premature or misaligned execution.
When & Why to Use This Skill
This Claude skill serves as a robust orchestrator for the 'prompt' skillset, designed to manage the lifecycle of prompt engineering and execution with high precision. By strictly separating intent discovery from final execution, it ensures a safe, predictable, and human-in-the-loop workflow. The skill leverages disk-based state management to maintain a single source of truth, preventing misaligned or premature AI actions while providing an auditable trail of execution receipts.
Use Cases
- Collaborative Prompt Engineering: Use the 'prompt-forge' capability to iteratively refine, stabilize, and store complex instructions on disk, ensuring the agent's intent is fully discovered and validated before any action is taken.
- Safe Execution of High-Stakes Tasks: Implement a 'deliberate execution' workflow where destructive or irreversible actions are only performed after a prompt artifact is finalized, protecting users from unintended AI behaviors.
- State-Consistent Agent Workflows: Manage complex agentic tasks in environments where conversation history is volatile by relying on authoritative disk state for persistent prompt management and auditability.
| name | prompt |
|---|---|
| description | > |
| author | Jordan Godau |
| version | 0.1.0 |
| schema_version | 1 |
| root | .prompt-forge |
| assets | [] |
| scripts | [] |
| references | [] |
| requires | [] |
Prompt Skillset
We forge prompts collaboratively, and we execute them deliberately.
Philosophy
This skillset assumes:
- Human intent is discovered, not known upfront
- Conversation is not state
- Disk state is authoritative
- Execution is irreversible and requires explicit consent
Core Invariants
1. Conversation is not state
The agent must not assume continuity or memory across turns. All durable state lives on disk. The agent may only read or modify state inside an explicit skill invocation.
Conversation is not state. Disk is state. Skills are the only legal state transitions.
2. Single canonical prompt on disk
There may be at most one active prompt artifact at any time.
Canonical path: .prompt-forge/active.yaml
3. Execution is destructive
After a successful prompt-exec:
- The canonical prompt artifact must be deleted
- The system must return to a "no active prompt" state
4. Optional execution receipts
To preserve auditability without violating single-prompt rules:
prompt-execmay write an execution receipt- Receipts are immutable and never executable
Receipt location: .prompt-forge/receipts/<timestamp>-<hash>.yaml
Member Skills
| Skill | Purpose |
|---|---|
prompt-forge |
Shape, refine, and stabilize intent into the canonical prompt artifact |
prompt-exec |
Execute the forged prompt exactly as written |
No other prompt-related skills are permitted.