<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: PRASEN JEET</title>
    <description>The latest articles on DEV Community by PRASEN JEET (@prasen-sky).</description>
    <link>https://dev.to/prasen-sky</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F739574%2F5d32076d-0cdf-408a-a079-0a2472bd5a6e.png</url>
      <title>DEV Community: PRASEN JEET</title>
      <link>https://dev.to/prasen-sky</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/prasen-sky"/>
    <language>en</language>
    <item>
      <title>I Built a Claude Code Skill That Reverse-Engineers Undocumented APIs</title>
      <dc:creator>PRASEN JEET</dc:creator>
      <pubDate>Wed, 19 Aug 2026 15:44:05 +0000</pubDate>
      <link>https://dev.to/prasen-sky/i-built-a-claude-code-skill-that-reverse-engineers-undocumented-apis-2h47</link>
      <guid>https://dev.to/prasen-sky/i-built-a-claude-code-skill-that-reverse-engineers-undocumented-apis-2h47</guid>
      <description>&lt;p&gt;I Built a Claude Code Skill That Reverse-Engineers Undocumented APIs&lt;/p&gt;

&lt;p&gt;Because "the docs are in the code" is not a documentation strategy.&lt;/p&gt;

&lt;p&gt;The Week I Lost to Grepping&lt;/p&gt;

&lt;p&gt;I joined a new team last month. Day 1 task: add a feature to the billing service.&lt;/p&gt;

&lt;p&gt;Day 1 reality: I opened the API docs and realized they were from 2022. Half the routes had been rewritten. The other half never had docs to begin with.&lt;/p&gt;

&lt;p&gt;So I did what every backend dev does. I grepped.&lt;/p&gt;

&lt;p&gt;grep -r "app.get|app.post|router." src/ --include="*.js"&lt;/p&gt;

&lt;p&gt;Four hours later, I had a notebook full of endpoints, a headache, and zero confidence that I had found everything.&lt;/p&gt;

&lt;p&gt;I found routes that worked but were not documented. I found docs for routes that did not exist anymore. I found one GET /invoices/:id endpoint with zero auth checks that had been sitting there since 2022.&lt;/p&gt;

&lt;p&gt;This is normal. And it should not be.&lt;/p&gt;

&lt;p&gt;The Idea&lt;/p&gt;

&lt;p&gt;What if I could drop a single file into a repo and have Claude Code map the entire API layer for me?&lt;/p&gt;

&lt;p&gt;Not from annotations. Not from existing OpenAPI specs. From the actual code.&lt;/p&gt;

&lt;p&gt;So I built it.&lt;/p&gt;

&lt;p&gt;Meet API Archaeologist&lt;/p&gt;

&lt;p&gt;API Archaeologist is a Claude Code / Codex CLI skill that reads your source code and reverse-engineers your API layer.&lt;/p&gt;

&lt;p&gt;It finds:&lt;/p&gt;

&lt;p&gt;• Internal endpoints — REST, GraphQL, gRPC, WebSockets&lt;br&gt;
• External integrations — third-party APIs, webhooks, SDK clients&lt;br&gt;
• Auth flows — JWT, OAuth, API keys, session cookies, RBAC&lt;br&gt;
• Security gaps — unauthenticated routes, hardcoded secrets, missing rate limits&lt;br&gt;
• Dead code — auth middleware with no endpoints, orphaned routes&lt;/p&gt;

&lt;p&gt;And it generates two things:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;API_DISCOVERY.md — A complete catalog with Mermaid diagrams&lt;/li&gt;
&lt;li&gt;openapi-draft.yaml — A draft OpenAPI spec&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;How It Works&lt;/p&gt;

&lt;p&gt;The skill is just a SKILL.md file. Claude Code reads it and follows the instructions.&lt;/p&gt;

&lt;p&gt;It:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Discovers route definitions&lt;/li&gt;
&lt;li&gt;Traces handlers, DTOs, middleware, services, and database calls&lt;/li&gt;
&lt;li&gt;Maps authentication and authorization&lt;/li&gt;
&lt;li&gt;Finds external API calls and integrations&lt;/li&gt;
&lt;li&gt;Flags potential security and reliability risks&lt;/li&gt;
&lt;li&gt;Generates API_DISCOVERY.md and openapi-draft.yaml&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Why Not Just Use Swagger?&lt;/p&gt;

&lt;p&gt;Swagger and OpenAPI Generator are great if your codebase already has annotations, decorators, or an existing specification.&lt;/p&gt;

&lt;p&gt;This is for the other kind of codebase:&lt;/p&gt;

&lt;p&gt;• Legacy monoliths&lt;br&gt;
• Startups without proper API documentation&lt;br&gt;
• Projects where the original developers have left&lt;br&gt;
• APIs that evolved faster than their documentation&lt;/p&gt;

&lt;p&gt;It reads the actual source code instead of depending on existing documentation.&lt;/p&gt;

&lt;p&gt;Installation&lt;/p&gt;

&lt;p&gt;Claude Code:&lt;/p&gt;

&lt;p&gt;mkdir -p ~/.claude/skills/api-archaeologist&lt;/p&gt;

&lt;p&gt;curl -o ~/.claude/skills/api-archaeologist/SKILL.md &lt;br&gt;
&lt;a href="https://raw.githubusercontent.com/prasen-sky/api-archaeologist/main/skills/api-archaeologist/SKILL.md" rel="noopener noreferrer"&gt;https://raw.githubusercontent.com/prasen-sky/api-archaeologist/main/skills/api-archaeologist/SKILL.md&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Codex CLI:&lt;/p&gt;

&lt;p&gt;mkdir -p ~/.codex/agents/skills/api-archaeologist&lt;/p&gt;

&lt;p&gt;curl -o ~/.codex/agents/skills/api-archaeologist/SKILL.md &lt;br&gt;
&lt;a href="https://raw.githubusercontent.com/prasen-sky/api-archaeologist/main/skills/api-archaeologist/SKILL.md" rel="noopener noreferrer"&gt;https://raw.githubusercontent.com/prasen-sky/api-archaeologist/main/skills/api-archaeologist/SKILL.md&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Or clone the repository:&lt;/p&gt;

&lt;p&gt;git clone &lt;a href="https://github.com/prasen-sky/api-archaeologist.git" rel="noopener noreferrer"&gt;https://github.com/prasen-sky/api-archaeologist.git&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;cp api-archaeologist/skills/api-archaeologist/SKILL.md ~/.claude/skills/api-archaeologist/&lt;/p&gt;

&lt;p&gt;Usage&lt;/p&gt;

&lt;p&gt;Navigate to your backend repository and run:&lt;/p&gt;

&lt;p&gt;claude /api-archaeologist&lt;/p&gt;

&lt;p&gt;Or with Codex:&lt;/p&gt;

&lt;p&gt;codex $api-archaeologist&lt;/p&gt;

&lt;p&gt;Then review the generated reports and verify the findings against your codebase.&lt;/p&gt;

&lt;p&gt;What I Learned&lt;/p&gt;

&lt;p&gt;The interesting part wasn't generating another API documentation tool.&lt;/p&gt;

&lt;p&gt;It was realizing how much useful information already exists inside a codebase.&lt;/p&gt;

&lt;p&gt;Routes, middleware, authentication, database calls, third-party APIs and request/response structures are already there.&lt;/p&gt;

&lt;p&gt;The problem is finding and connecting all of it.&lt;/p&gt;

&lt;p&gt;Limitations&lt;/p&gt;

&lt;p&gt;The OpenAPI output is a draft. Types and behavior may need manual verification.&lt;/p&gt;

&lt;p&gt;Dynamic or heavily meta-programmed routing can be harder to analyze.&lt;/p&gt;

&lt;p&gt;Large monorepos are better analyzed service by service.&lt;/p&gt;

&lt;p&gt;Roadmap&lt;/p&gt;

&lt;p&gt;• Frontend API consumer mapping&lt;br&gt;
• Postman collection export&lt;br&gt;
• CI/CD integration&lt;br&gt;
• Detection of newly introduced unauthenticated endpoints&lt;/p&gt;

&lt;p&gt;Try It&lt;/p&gt;

&lt;p&gt;If your API documentation is outdated, this might be useful.&lt;/p&gt;

&lt;p&gt;GitHub:&lt;br&gt;
[&lt;a href="https://github.com/prasen-sky/api-archaeologist" rel="noopener noreferrer"&gt;https://github.com/prasen-sky/api-archaeologist&lt;/a&gt;]&lt;/p&gt;

&lt;p&gt;Try it on a repo and let me know what it finds.&lt;/p&gt;

</description>
      <category>ai</category>
      <category>webdev</category>
      <category>productivity</category>
      <category>api</category>
    </item>
  </channel>
</rss>
