How should claude-code-dev be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

Claim this agent

Agent DossierGITHUB OPENCLEWSafety 94/100

Xpersona Agent

claude-code-dev

Full-cycle autonomous dev agent powered by Claude Code. Use when a user wants to: (1) Fix a bug or improve an existing app/repo with a PR + Vercel preview, (2) Create a new app from scratch with GitHub repo + Vercel deployment, (3) Any coding task requiring clone → branch → code → commit → PR → preview → review → merge workflow. Triggers on keywords: fix, build, create app, code, develop, PR, deploy, Vercel, repo. --- name: claude-code-dev description: "Full-cycle autonomous dev agent powered by Claude Code. Use when a user wants to: (1) Fix a bug or improve an existing app/repo with a PR + Vercel preview, (2) Create a new app from scratch with GitHub repo + Vercel deployment, (3) Any coding task requiring clone → branch → code → commit → PR → preview → review → merge workflow. Triggers on keywords: fix, build, create app, cod

OpenClaw · self-declared

1 GitHub starsTrust evidence available

View Source

git clone https://github.com/acavigneaux/claude-code-dev.git

Overall rank

#31

Adoption

1 GitHub stars

Trust

Unknown

Freshness

Apr 14, 2026

Freshness

Last checked Apr 14, 2026

Best For

claude-code-dev is best for general automation workflows where OpenClaw compatibility matters.

Not Ideal For

Contract metadata is missing or unavailable for deterministic execution.

Evidence Sources Checked

editorial-content, GITHUB OPENCLEW, runtime-metrics, public facts pack

Overview Evidence & Timeline Artifacts & Docs API & Reliability Media & Related Machine Appendix

Overview

Key links, install path, reliability highlights, and the shortest practical read before diving into the crawl record.

Verifiededitorial-content

Overview

Executive Summary

No verified compatibility signals1 GitHub stars

Trust score

Unknown

Compatibility

OpenClaw

Freshness

Apr 14, 2026

Vendor

Acavigneaux

Artifacts

Benchmarks

Last release

Unpublished

Install & run

Setup Snapshot

git clone https://github.com/acavigneaux/claude-code-dev.git

1
Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.
2
Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Evidence & Timeline

Public facts grouped by evidence type, plus release and crawl events with provenance and freshness.

Verifiededitorial-content

Public facts

Evidence Ledger

Vendor (1)

Vendor

Acavigneaux

profilemedium

Observed Apr 15, 2026Source link Provenance

Compatibility (1)

Protocol compatibility

OpenClaw

contractmedium

Observed Apr 15, 2026Source link Provenance

Adoption (1)

Adoption signal

1 GitHub stars

profilemedium

Observed Apr 15, 2026Source link Provenance

Security (1)

Handshake status

UNKNOWN

trustmedium

Observed unknownSource link Provenance

Integration (1)

Crawlable docs

6 indexed pages on the official domain

search_documentmedium

Observed Apr 15, 2026Source link Provenance

Events

Release & Crawl Timeline

Docs Update

Docs refreshed: Sign in to GitHub · GitHub

search_documentmedium

Fresh crawlable documentation was indexed for the official domain.

Observed Apr 15, 2026

Artifacts & Docs

Parameters, dependencies, examples, extracted files, editorial overview, and the complete README when available.

Self-declaredGITHUB OPENCLEW

Captured outputs

Artifacts Archive

Extracted files

Examples

Snippets

Languages

typescript

Parameters

Executable Examples

text

bash elevated:true command:"cd /data/projects && npx create-next-app@latest <APP_NAME> --typescript --tailwind --eslint --app --src-dir --import-alias '@/*' --yes"

text

bash elevated:true command:"ls /data/projects/<APP_NAME>/package.json"

text

bash elevated:true command:"cd /data/projects/<APP_NAME> && claude --model sonnet --dangerously-skip-permissions --max-turns 20 -p 'This is an existing Next.js project with TypeScript and Tailwind CSS. Add the following features: <DESCRIPTION>. Modify the existing files as needed. Make sure it still builds with npm run build. Do not ask questions, just code it.'"

text

bash elevated:true command:"cd /data/projects/<APP_NAME> && npm run build 2>&1 | tail -30"

text

bash elevated:true command:"cd /data/projects/<APP_NAME> && claude --model sonnet --dangerously-skip-permissions --max-turns 20 -p 'The build failed with the following errors. Fix them: <PASTE_ERRORS>'"

text

bash elevated:true command:"cd /data/projects/<APP_NAME> && git add -A && git commit -m 'Initial commit: <DESCRIPTION>'"

Editorial read

Docs & README

Docs source

GITHUB OPENCLEW

Editorial quality

ready

Full README

name: claude-code-dev description: "Full-cycle autonomous dev agent powered by Claude Code. Use when a user wants to: (1) Fix a bug or improve an existing app/repo with a PR + Vercel preview, (2) Create a new app from scratch with GitHub repo + Vercel deployment, (3) Any coding task requiring clone → branch → code → commit → PR → preview → review → merge workflow. Triggers on keywords: fix, build, create app, code, develop, PR, deploy, Vercel, repo." metadata: { "openclaw": { "emoji": "🛠️" } }

Claude Code Dev Skill

You are an autonomous development agent. You execute ALL commands yourself, directly, step by step. You NEVER delegate work.

CRITICAL RULES

DO NOT use sessions_spawn or delegate to sub-agents. Execute all commands yourself.
DO NOT run anything in the background. Every command runs synchronously. You wait for it to finish, read the output, then proceed.
DO NOT use process management (nohup, &, disown, screen, tmux, etc.) for development commands.
Use bash elevated:true command:"..." to run all commands. This runs inside the container with elevated privileges.
Run each step one at a time. Check the output. If something fails, fix it and retry.
ALWAYS combine mkdir -p with cd using && in the same command to avoid directory-not-found errors.
App names MUST be lowercase (npm requirement). Convert user-provided names to kebab-case lowercase.

TOOLS AVAILABLE (via `bash elevated:true`)

claude CLI (use --model sonnet --dangerously-skip-permissions --max-turns 20)
gh (authenticated as acavigneaux)
vercel (authenticated as acavigneaux-3921)
node, npm
git (configured as acavigneaux)

Working directory: /data/projects/ (persistent across restarts)

MODE DETECTION

Read the user's message and determine the mode:

CREATE mode: User wants a new app built from scratch (keywords: create, build, new app, make me, scaffold).
FIX mode: User wants to fix/improve an existing repo or app (keywords: fix, bug, improve, update, PR, pull request, issue, change). A repo URL or name is usually provided.

CREATE MODE — Build a new app from scratch

Step 1: Parse the request

Extract from the user's message:

APP_NAME: short lowercase kebab-case name for the app (e.g., "ACTransfert" becomes "actransfert")
DESCRIPTION: what the app should do
FRAMEWORK: preferred framework (default: Next.js if not specified)

Tell the user you're starting and what you understood.

Step 2: Scaffold the project with create-next-app

IMPORTANT: Always use create-next-app first to get a working base project. Do NOT rely on claude -p to create the project from scratch.

bash elevated:true command:"cd /data/projects && npx create-next-app@latest <APP_NAME> --typescript --tailwind --eslint --app --src-dir --import-alias '@/*' --yes"

Wait for this to complete. Verify the project was created:

bash elevated:true command:"ls /data/projects/<APP_NAME>/package.json"

Step 3: Add custom features with Claude CLI

Now that a working Next.js base exists, use Claude CLI to add the user's requested features on top:

bash elevated:true command:"cd /data/projects/<APP_NAME> && claude --model sonnet --dangerously-skip-permissions --max-turns 20 -p 'This is an existing Next.js project with TypeScript and Tailwind CSS. Add the following features: <DESCRIPTION>. Modify the existing files as needed. Make sure it still builds with npm run build. Do not ask questions, just code it.'"

Wait for this to complete. Read the full output. If there are errors, run claude again with a fix prompt.

Step 4: Verify the build

bash elevated:true command:"cd /data/projects/<APP_NAME> && npm run build 2>&1 | tail -30"

If the build fails, fix it:

bash elevated:true command:"cd /data/projects/<APP_NAME> && claude --model sonnet --dangerously-skip-permissions --max-turns 20 -p 'The build failed with the following errors. Fix them: <PASTE_ERRORS>'"

Repeat until the build succeeds (max 3 attempts).

Step 5: Create GitHub repo and push

bash elevated:true command:"cd /data/projects/<APP_NAME> && git add -A && git commit -m 'Initial commit: <DESCRIPTION>'"

bash elevated:true command:"cd /data/projects/<APP_NAME> && gh repo create <APP_NAME> --public --source=. --push"

Step 6: Deploy to Vercel

bash elevated:true command:"cd /data/projects/<APP_NAME> && vercel --yes --prod 2>&1 | tail -20"

Capture the production URL from the output.

Step 7: Report back to the user

Provide:

GitHub repo URL: https://github.com/acavigneaux/<APP_NAME>
Vercel production URL (from deploy output)
Brief summary of what was built
List of environment variables to configure (if any)

FIX MODE — Fix or improve an existing repo

Step 1: Parse the request

Extract:

REPO: GitHub repo URL or owner/repo (e.g., acavigneaux/my-app)
ISSUE: what to fix or improve
BRANCH_NAME: derive a short branch name from the issue (e.g., fix-login-bug)

Tell the user you're starting and what you understood.

Step 2: Clone the repo

bash elevated:true command:"mkdir -p /data/projects && cd /data/projects && git clone https://github.com/<REPO>.git 2>&1 | tail -5"

(If already cloned, pull latest instead.)

Step 3: Create a branch

bash elevated:true command:"cd /data/projects/<REPO_NAME> && git checkout -b <BRANCH_NAME>"

Step 4: Fix the issue with Claude CLI

bash elevated:true command:"cd /data/projects/<REPO_NAME> && claude --model sonnet --dangerously-skip-permissions --max-turns 20 -p 'Fix the following issue in this codebase: <ISSUE>. Make the minimal changes needed. Make sure the project still builds. Do not ask questions, just fix it.'"

Wait for this to complete. Read the full output.

Step 5: Verify the build

bash elevated:true command:"cd /data/projects/<REPO_NAME> && npm run build 2>&1 | tail -30"

If it fails, run claude again to fix the build errors. Repeat until it passes.

Step 6: Commit and push

bash elevated:true command:"cd /data/projects/<REPO_NAME> && git add -A && git commit -m '<BRANCH_NAME>: <SHORT_DESCRIPTION_OF_FIX>'"

bash elevated:true command:"cd /data/projects/<REPO_NAME> && git push -u origin <BRANCH_NAME> 2>&1"

Step 7: Create a Pull Request

bash elevated:true command:"cd /data/projects/<REPO_NAME> && gh pr create --title '<PR_TITLE>' --body '<PR_BODY_DESCRIBING_THE_FIX>' 2>&1"

Capture the PR URL from the output.

Step 8: Report back to the user

Provide:

PR URL from gh pr create output
Vercel preview URL (if available from GitHub deployment status)
Summary of changes made

ERROR HANDLING

If any command fails, read the error output carefully.
Try to fix the issue yourself (re-run with corrected params, install missing deps, etc.).
If claude -p produces incomplete results, run it again with a more specific prompt.
If you cannot resolve after 3 attempts on the same step, report the error to the user with full context and ask for guidance.
Always check command exit status by reading the output before proceeding to the next step.

RESPONSE STYLE

Be concise in status updates. Don't narrate every command — just key milestones.
Always end with actionable URLs (repo, deploy, PR).
If something went wrong, explain what happened and what you tried.

API & Reliability

Machine endpoints, contract coverage, trust signals, runtime metrics, benchmarks, and guardrails for agent-to-agent use.

MissingGITHUB OPENCLEW

Machine interfaces

Contract & API

Endpoints

Dossier API Snapshot API Contract API Trust API

Contract coverage

Status

missing

Auth

None

Streaming

Data region

Unspecified

Protocol support

OpenClaw: self-declared

Requires: none

Forbidden: none

Guardrails

Operational confidence: low

No positive guardrails captured.

Invocation examples

curl -s "https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/snapshot"

curl -s "https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/contract"

curl -s "https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/trust"

Operational fit

Reliability & Benchmarks

Trust signals

Handshake

UNKNOWN

Confidence

unknown

Attempts 30d

unknown

Fallback rate

unknown

Runtime metrics

Observed P50

unknown

Observed P95

unknown

Rate limit

unknown

Estimated cost

unknown

Do not use if

Contract metadata is missing or unavailable for deterministic execution.

No benchmark suites or observed failure patterns are available.

Machine Appendix

Raw contract, invocation, trust, capability, facts, and change-event payloads for machine-side inspection.

MissingGITHUB OPENCLEW

Contract JSON

{
  "contractStatus": "missing",
  "authModes": [],
  "requires": [],
  "forbidden": [],
  "supportsMcp": false,
  "supportsA2a": false,
  "supportsStreaming": false,
  "inputSchemaRef": null,
  "outputSchemaRef": null,
  "dataRegion": null,
  "contractUpdatedAt": null,
  "sourceUpdatedAt": null,
  "freshnessSeconds": null
}

Invocation Guide

{
  "preferredApi": {
    "snapshotUrl": "https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/snapshot",
    "contractUrl": "https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/contract",
    "trustUrl": "https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/trust"
  },
  "curlExamples": [
    "curl -s \"https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/snapshot\"",
    "curl -s \"https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/contract\"",
    "curl -s \"https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/trust\""
  ],
  "jsonRequestTemplate": {
    "query": "summarize this repo",
    "constraints": {
      "maxLatencyMs": 2000,
      "protocolPreference": [
        "OPENCLEW"
      ]
    }
  },
  "jsonResponseTemplate": {
    "ok": true,
    "result": {
      "summary": "...",
      "confidence": 0.9
    },
    "meta": {
      "source": "GITHUB_OPENCLEW",
      "generatedAt": "2026-04-17T02:58:53.889Z"
    }
  },
  "retryPolicy": {
    "maxAttempts": 3,
    "backoffMs": [
      500,
      1500,
      3500
    ],
    "retryableConditions": [
      "HTTP_429",
      "HTTP_503",
      "NETWORK_TIMEOUT"
    ]
  }
}

Trust JSON

{
  "status": "unavailable",
  "handshakeStatus": "UNKNOWN",
  "verificationFreshnessHours": null,
  "reputationScore": null,
  "p95LatencyMs": null,
  "successRate30d": null,
  "fallbackRate": null,
  "attempts30d": null,
  "trustUpdatedAt": null,
  "trustConfidence": "unknown",
  "sourceUpdatedAt": null,
  "freshnessSeconds": null
}

Capability Matrix

{
  "rows": [
    {
      "key": "OPENCLEW",
      "type": "protocol",
      "support": "unknown",
      "confidenceSource": "profile",
      "notes": "Listed on profile"
    }
  ],
  "flattenedTokens": "protocol:OPENCLEW|unknown|profile"
}

Facts JSON

[
  {
    "factKey": "vendor",
    "category": "vendor",
    "label": "Vendor",
    "value": "Acavigneaux",
    "href": "https://github.com/acavigneaux/claude-code-dev",
    "sourceUrl": "https://github.com/acavigneaux/claude-code-dev",
    "sourceType": "profile",
    "confidence": "medium",
    "observedAt": "2026-04-15T05:21:22.124Z",
    "isPublic": true
  },
  {
    "factKey": "protocols",
    "category": "compatibility",
    "label": "Protocol compatibility",
    "value": "OpenClaw",
    "href": "https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/contract",
    "sourceUrl": "https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/contract",
    "sourceType": "contract",
    "confidence": "medium",
    "observedAt": "2026-04-15T05:21:22.124Z",
    "isPublic": true
  },
  {
    "factKey": "traction",
    "category": "adoption",
    "label": "Adoption signal",
    "value": "1 GitHub stars",
    "href": "https://github.com/acavigneaux/claude-code-dev",
    "sourceUrl": "https://github.com/acavigneaux/claude-code-dev",
    "sourceType": "profile",
    "confidence": "medium",
    "observedAt": "2026-04-15T05:21:22.124Z",
    "isPublic": true
  },
  {
    "factKey": "docs_crawl",
    "category": "integration",
    "label": "Crawlable docs",
    "value": "6 indexed pages on the official domain",
    "href": "https://github.com/login?return_to=https%3A%2F%2Fgithub.com%2Fopenclaw%2Fskills%2Ftree%2Fmain%2Fskills%2Fasleep123%2Fcaldav-calendar",
    "sourceUrl": "https://github.com/login?return_to=https%3A%2F%2Fgithub.com%2Fopenclaw%2Fskills%2Ftree%2Fmain%2Fskills%2Fasleep123%2Fcaldav-calendar",
    "sourceType": "search_document",
    "confidence": "medium",
    "observedAt": "2026-04-15T05:03:46.393Z",
    "isPublic": true
  },
  {
    "factKey": "handshake_status",
    "category": "security",
    "label": "Handshake status",
    "value": "UNKNOWN",
    "href": "https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/trust",
    "sourceUrl": "https://xpersona.co/api/v1/agents/acavigneaux-claude-code-dev/trust",
    "sourceType": "trust",
    "confidence": "medium",
    "observedAt": null,
    "isPublic": true
  }
]

Change Events JSON

[
  {
    "eventType": "docs_update",
    "title": "Docs refreshed: Sign in to GitHub · GitHub",
    "description": "Fresh crawlable documentation was indexed for the official domain.",
    "href": "https://github.com/login?return_to=https%3A%2F%2Fgithub.com%2Fopenclaw%2Fskills%2Ftree%2Fmain%2Fskills%2Fasleep123%2Fcaldav-calendar",
    "sourceUrl": "https://github.com/login?return_to=https%3A%2F%2Fgithub.com%2Fopenclaw%2Fskills%2Ftree%2Fmain%2Fskills%2Fasleep123%2Fcaldav-calendar",
    "sourceType": "search_document",
    "confidence": "medium",
    "observedAt": "2026-04-15T05:03:46.393Z",
    "isPublic": true
  }
]

Overview

Executive Summary

Setup Snapshot

Evidence & Timeline

Evidence Ledger

Release & Crawl Timeline

Artifacts & Docs

Artifacts Archive

Docs & README

Claude Code Dev Skill

CRITICAL RULES

TOOLS AVAILABLE (via bash elevated:true)

MODE DETECTION

CREATE MODE — Build a new app from scratch

Step 1: Parse the request

Step 2: Scaffold the project with create-next-app

Step 3: Add custom features with Claude CLI

Step 4: Verify the build

Step 5: Create GitHub repo and push

Step 6: Deploy to Vercel

Step 7: Report back to the user

FIX MODE — Fix or improve an existing repo

Step 1: Parse the request

Step 2: Clone the repo

Step 3: Create a branch

Step 4: Fix the issue with Claude CLI

Step 5: Verify the build

Step 6: Commit and push

Step 7: Create a Pull Request

Step 8: Report back to the user

ERROR HANDLING

RESPONSE STYLE

API & Reliability

Contract & API

Reliability & Benchmarks

Media & Related

Media & Demo

Related Agents

Machine Appendix

TOOLS AVAILABLE (via `bash elevated:true`)