ClawHub

OpenClaw Pilot

v0.3.0-beta.2

Convert rough project ideas into a human-readable blueprint and a clean OpenClaw execution packet for safe, scoped implementation and continuation.

⭐ 1· 101·0 current·0 all-time
by@catcherintheroad-hub

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for catcherintheroad-hub/openclaw-pilot.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "OpenClaw Pilot" (catcherintheroad-hub/openclaw-pilot) from ClawHub.
Skill page: https://clawhub.ai/catcherintheroad-hub/openclaw-pilot
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install openclaw-pilot

ClawHub CLI

Package manager switcher

npx clawhub@latest install openclaw-pilot
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name and description match the actual requirements and behavior: no binaries, no env vars, no installs, and the SKILL.md only defines a two-message planning/packet contract.
✓
Instruction Scope
SKILL.md directs the agent only to produce a human-facing blueprint and a separate, exact execution-packet message; it does not instruct reading files, accessing unrelated env vars, or contacting external endpoints.
✓
Install Mechanism
No install spec or code files are present (instruction-only skill), so nothing is written to disk or fetched at install time.
✓
Credentials
The skill requests no environment variables, credentials, or config paths — proportional for an instruction-only planning skill.
✓
Persistence & Privilege
always is false and there is no indication the skill modifies agent/system configuration or requests persistent privileges.
Assessment
This skill appears coherent and low-risk as an instruction-only planner. Before handing any generated execution packet to OpenClaw (or any automation), review the packet contents for scope, safety, and unintended commands — the skill guarantees formatting but not that the packet itself is safe. If you plan to integrate this with an automated executor, enforce additional validation and least-privilege controls on what execution packets are allowed to do.

Like a lobster shell, security has layers — review code before you run it.

latestvk972sn06a9nn19qnkfqqjzqc0984763z
101downloads
1stars
2versions
Updated 3w ago
v0.3.0-beta.2
MIT-0
@catcherintheroad-hub
latest
Download

OpenClaw Pilot

Compile rough ideas into an executable blueprint and a clean OpenClaw execution packet.

What this skill does

OpenClaw Pilot is an instruction-first skill built around one core value: compile rough ideas into an executable blueprint and a clean OpenClaw execution packet.

It helps the user convert a fuzzy goal into two tightly scoped outputs:

  1. a human-readable blueprint for the current stage
  2. a separate pure execution packet that can be handed directly to OpenClaw

It also supports continuation through /pilot next <pilot_id> ..., so the user can keep advancing the same pilot without re-explaining the whole project.

Core output contract

This skill is built around a strict two-message delivery contract.

Message 1 — human-readable blueprint

The first message is for the human. It should explain:

  • the current goal
  • the current stage
  • why this stage is the right next step
  • in-scope vs out-of-scope boundaries
  • what feedback should come back to /pilot
  • the next continuation command

Message 2 — pure execution packet

The second message must contain only one block:

[OPENCLAW_EXECUTION_PACKET v1]
...
[END_OPENCLAW_EXECUTION_PACKET]

No extra prose. No prefix. No suffix. No human commentary mixed into the packet message.

Use when

Use OpenClaw Pilot when the user:

  • has a rough project idea but needs it compiled into an executable plan
  • wants a safe, scoped blueprint before execution
  • wants a clean packet to pass into OpenClaw
  • needs to continue an existing pilot with /pilot next <pilot_id> ...
  • wants the behavior contract locked down more than host-specific integration tricks

Do not use for

  • unrelated host integration work
  • broad plugin-platform refactors when a skill shell is enough
  • pretending to execute work that has not been scoped
  • merging human-readable explanation into the packet message

Command patterns

New pilot

/pilot <goal>

Example:

/pilot Build a lightweight OpenClaw workflow for turning vague product ideas into implementation-ready execution packets.

Continue an existing pilot

/pilot next <pilot_id> <feedback>

Example:

/pilot next pilot-abc123 STATUS: blocked SUMMARY: docs are clear but the packet needs stricter scope boundaries.

Behavior promises for the public beta

This first public beta skill release promises:

  • blueprint-first planning
  • strict two-message output contract
  • clean packet handoff semantics
  • continuation with /pilot next <pilot_id> ...
  • instruction-first usage with minimal setup burden

It does not promise:

  • a heavy plugin companion
  • remote host automation as the headline feature
  • broad environment/bootstrap orchestration
  • perfect immunity to every upstream model-formatting wobble

Notes for publishers

This skill is intentionally metadata-clean and dependency-light.

  • Prefer publishing the skill shell first.
  • Keep examples concrete and contract-focused.
  • Do not declare extra local binaries unless they are truly required.
  • If a future plugin companion is added, keep its boundary separate from this skill’s public contract.

Comments

Loading comments...