Along Plan for Openclaw

A read-only exploration and planning protocol that enforces safe code analysis before any modifications are made.

orangon
v1.0.0
Mar 31, 2026
0
586
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install along-plan

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install along-plan using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Along Plan?

The Along Plan skill is a specialized protocol designed for Openclaw Skills that prioritizes safety and structured planning. It transitions the AI agent into a dedicated plan mode where it is restricted to read-only tools, preventing accidental or premature code changes. By focusing on exploration through grep, glob, and safe bash commands, it ensures the agent fully understands the codebase before proposing an execution strategy.

This skill creates a clear boundary between analysis and implementation. It requires the agent to produce a formalized plan document, which serves as a roadmap for subsequent tasks. By utilizing Openclaw Skills like Along Plan, developers can ensure that complex refactors or features are meticulously mapped out and validated against acceptance criteria before a single line of functional code is altered.

Along Plan Use Cases

  • Entering a dedicated plan mode for complex architectural changes.
  • Conducting deep code analysis without the risk of making accidental edits.
  • Creating a documented roadmap for multi-step tasks.
  • Validating technical assumptions using safe search and discovery tools.
  • Enforcing a structured hand-off between analysis and execution phases.

How Along Plan Works

  1. The agent enters plan mode upon user request or when a complex task is identified.
  2. Access is restricted to read-only tools such as read, grep, and glob, along with a safe subset of bash commands.
  3. The agent explores the codebase to gather context and identify necessary changes.
  4. A numbered list of steps is generated under a specific Plan: header.
  5. The finalized plan is saved to a markdown file in the documentation directory, including TODO items and specific acceptance criteria.

Along Plan Setup

To utilize Along Plan within your Openclaw Skills environment, ensure the skill definition is present in your skill library. You must also have a references/safe-commands.md file defined to govern the bash tool restrictions.

# Ensure documentation directory exists
mkdir -p docs

# The agent will automatically trigger plan mode when asked to "create a plan"

Ensure either a docs/ or doc/ directory exists in your project root for plan storage.

Along Plan Data Schema & Taxonomy

Feature Description
Plan Header Requires an exact Plan: prefix for tool discovery.
Storage Path Saves to docs/plan-.md or doc/plan-.md.
Documentation Format Markdown with # Plan, ## TODO, and ## Acceptance Criteria sections.
Tracking Uses [DONE:n] markers to track step completion during the execution phase.

Along Plan Advanced Features

  • Enforced tool restriction (read-only mode) for maximum safety during exploration.
  • Integration with safe-commands.md for granular bash command filtering within Openclaw Skills.
  • Automatic plan persistence to project documentation for audit trails and team collaboration.
  • Structured acceptance criteria definitions to ensure successful task outcomes.
  • Explicit completion tracking with standardized markers for multi-step execution.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*