Skip to content
Back to skills

Amp

ASecurity

Reference system prompt and tool schemas for Amp. Use when imitating, comparing, or analyzing Amp's prompting style; designing prompts inspired by Amp; or answering questions about how Amp is prompted.

  • 666 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added October 8, 2026
ai-agents

Security analysis

A100/100

Scanned October 8, 2026

npx -y skills add majiayu000/claude-skill-registry --skill amp --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Amp?

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

Security grade badge for Amp
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/majiayu000-amp/badge)](https://www.skillsdirectory.com/skills/majiayu000-amp)

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

SKILL.md
---
name: amp
description: Reference system prompt and tool schemas for Amp. Use when imitating, comparing, or analyzing Amp's prompting style; designing prompts inspired by Amp; or answering questions about how Amp is prompted.
---

# Amp

Sourcegraph Amp coding agent.

When activated, read the bundled source files below to ground your response
in the actual published prompt text. Do not paraphrase from memory — the
canonical content lives in these files.

## Source files

- `sources/README.md` (614 bytes)
- `sources/claude-4-sonnet.yaml` (66,230 bytes)
- `sources/gpt-5.yaml` (62,239 bytes)

## Usage

- Quote or summarise from the files above when answering "how does Amp prompt its agent?" style questions.
- When emulating Amp's style, mirror its structure (sections, tool-use rules, refusal patterns) rather than copying verbatim.
- These prompts were collected by the upstream `x1xhlol/system-prompts-and-models-of-ai-tools` repo and may drift from the live product over time.

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.