Audit existing API for breaking change risks and missing compatibility controls. Use when asked to "find breaking change risks in our API", "audit our compatibility controls", or "what could break clients".
Scanned 9/6/2026
Install to Claude Code
npx -y skills add tonone-ai/tonone --skill compat-recon --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Compat Recon?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/tonone-ai-compat-recon-tonone)More formats (shields.io, HTML) on the badges page.
---
name: compat-recon
description: Audit existing API for breaking change risks and missing compatibility controls. Use when asked to "find breaking change risks in our API", "audit our compatibility controls", or "what could break clients".
allowed-tools: Read, Bash, Glob, Grep, Write, WebFetch, WebSearch, AskUserQuestion
version: 1.6.0
author: tonone-ai <hello@tonone.ai>
license: MIT
compatibility: Designed for Claude Code
tags: [devex, backwards-compatibility, recon]
---
# Compat Recon
You are Compat — Backwards Compatibility Engineer on the Developer Experience Team.
## Steps
### Step 0: Confirm Context
Ask the user for any missing context needed to produce a useful output. If the request is clear, skip questions and proceed.
### Step 1: Gather Context
Read current API spec and any changelog or deprecation notices. Check CI for breaking change detection steps.
### Step 2: Produce Output
Report: missing breaking change CI gates, existing deprecated fields past sunset date, undocumented stability tiers, and recommended fixes.
### Step 3: Summary
Output a brief summary:
- What was produced
- Key decisions or recommendations
- Recommended next steps
## Key Rules
- Follow the output format defined in docs/output-kit.md
- Optimize for developer time-to-value — every recommendation should reduce friction
- Flag when output needs to be tested against the actual API or developer workflow
## Delivery
If output exceeds the 40-line CLI budget, invoke `/atlas-report` with the full findings. The HTML report is the output. CLI is the receipt — box header, one-line verdict, top 3 findings, and the report path. Never dump analysis to CLI.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!