read-prompt-file

Load a branch-aware prompt file into the active Claude conversation.

Updated May 24, 2026
One-click install
npx skills add https://github.com/empty-can/cc-relative-info --skill read-prompt-file
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: read-prompt-file
Source: https://github.com/empty-can/cc-relative-info/tree/main/.claude/skills/read-prompt-file
Command: npx skills add https://github.com/empty-can/cc-relative-info --skill read-prompt-file

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps when you want to write long or multi-line prompts in your own editor instead of the chat input, and then pass that saved text into Claude as working context. It also preserves the prompt file after use so you can review or reuse what you sent later.

Core Features & Use Cases

  • Branch-aware prompt discovery: Resolves the correct .claude/work/prompt.txt location based on the current branch's activity folder rules.
  • Clear missing-file handling: Stops immediately with a fixed guidance message when the prompt file does not exist.
  • Line-numbered prompt loading: Reads the saved prompt file with line numbers so the content can be inspected and processed accurately.
  • Use case: Draft a complex review request or implementation brief in your editor, save it to the branch-specific prompt file, and invoke the Skill to continue the conversation from that saved text.

Quick Start

Save your message in the appropriate .claude/work/prompt.txt file for the current branch and ask Claude to run the read-prompt-file skill.

Frequently Asked Questions about read-prompt-file

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I load a saved prompt file into Claude Code from my workspace?

To load a saved prompt file into Claude Code, this Skill resolves the branch-aware workspace path and injects the text contents directly into your active conversation. It reads the file with line numbers for accurate context processing.

Can I draft multi-line prompts in my editor and pass them to Claude Code?

Yes, you can draft multi-line prompts in your editor and pass them to Claude Code. Save your message to the branch-specific prompt file, and this Skill will load that text into the active conversation as working context.

What is the best way to reuse long-form instruction prompts across different Git branches?

The best way to reuse long-form prompts across branches is using branch-aware prompt discovery. This Skill resolves the correct branch-specific workspace path to load the appropriate saved prompt file for your current context.

Does this prompt loading tool work with Bash-based path resolution in Claude Code?

Yes, this prompt loading tool uses Bash-based path resolution to locate the prompt file within the Claude Code repository workspace. It requires Bash to validate file existence and read the line-numbered prompt contents.

What happens if the prompt file does not exist in the Claude workspace path?

If the prompt file does not exist in the Claude workspace path, the Skill stops immediately and provides a fixed guidance message. This missing-file handling ensures you are alerted before attempting to inject empty context.

Are loaded prompt files deleted from the workspace after being injected into Claude?

Loaded prompt files are not deleted after being injected into Claude. The Skill preserves the prompt file in your branch workspace so you can review or reuse the sent text later for future conversations.