Skills clawdirect-dev
📦
clawdirect-dev
Content revision r1 High Risk ⚡ Contains scripts⚙️ External commands🌐 Network access🔑 Env variables
Build ATXP Agent Web Apps
Agent-facing web apps need reliable authentication between MCP tools and browsers. This skill provides templates for ATXP auth, cookie setup, Express APIs, and directory publishing.
Supports: Claude Codex Code(CC)
Stop for confirmation before installing.
Review the plan and obtain explicit user consent before changing files.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Agent request
Review the Skillstore skill "clawdirect-dev" from https://skillstore.io/skills/napoleond-clawdirect-dev.md and its manifest at https://skillstore.io/api/skills/napoleond-clawdirect-dev/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "clawdirect-dev". Plan an ATXP web app for a document review service.
Expected outcome:
A project plan with MCP tools, browser login flow, protected API routes, storage choices, and deployment notes.
Using "clawdirect-dev". Add browser authentication for agents that already call my MCP server.
Expected outcome:
A step-by-step implementation outline for issuing cookies, validating requests, and redirecting agents after authentication.
Using "clawdirect-dev". Prepare my app for ClawDirect publication.
Expected outcome:
A publishing checklist covering the listing fields, thumbnail requirements, ownership rules, costs, and deletion behavior.
Security Audit
High Riskv4 • 7/5/2026 Open versioned report
Most static hits are Markdown fences, TypeScript template literals, configuration examples, or reference links rather than hidden execution. The review confirms some explicit setup and npx network commands, plus a high-risk auth design that passes bearer cookies through URLs. The skill should not be treated as production-ready until cookie transport and token storage are hardened.
1
Files scanned
501
Lines analyzed
12
Review items
0
False positives ignored
Confirmed security concerns (3)
High
Authentication Cookie Passed Through Query String
The skill recommends passing a browser authentication cookie as a URL query parameter and then redirecting. Bearer tokens in URLs can leak through server logs, browser history, analytics tools, screenshots, and referrer headers.
Multiple sections explicitly recommend the query-string cookie bootstrap. This is a known bearer-token exposure pattern even though it is framed as an agent browser workaround.
Medium
SQLite database file
const DB_PATH = process.env.DB_PATH || './data.db';
The template stores authentication cookie records in a local SQLite database. This is a legitimate design, but raw bearer token storage needs stronger handling and review.
Medium
Long-Lived Bearer Cookies Stored Without Expiry Enforcement
The template stores raw cookie values in SQLite and sets a 30-day browser cookie, but the table schema and lookup flow do not enforce expiration, hashing, rotation, or revocation. A database disclosure could expose reusable authentication tokens.
The schema stores cookie_value directly and records only created_at. The validation function checks token equality without expiration or revocation logic.
Capability review items (12)
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
High
Hardcoded URL
instructions: 'To authenticate in a browser, navigate to https://your-domain.com?myapp_cookie=<cooki
The URL demonstrates passing an authentication cookie in the query string. That can expose a bearer token through logs, browser history, analytics, and referrer headers.
High
Hardcoded URL
https://your-domain.com?myapp_cookie=<cookie_value>
The URL demonstrates passing an authentication cookie in the query string. That can expose a bearer token through logs, browser history, analytics, and referrer headers.
Medium
Ruby/shell backtick execution
```bash
This fenced bash setup asks users or agents to run mkdir, npm init, and npm install commands. The commands are visible and relevant, but they install external packages and change the filesystem.
Medium
Ruby/shell backtick execution
1. Install ATXP: `npx skills add atxp-dev/cli --skill atxp`
This template includes npx commands that install or call ATXP tooling over the network. The commands are purposeful but should require user confirmation before execution.
Medium
Ruby/shell backtick execution
2. Call MCP tools: `npx atxp-call https://your-domain.com/mcp <tool> [params]`
This template includes npx commands that install or call ATXP tooling over the network. The commands are purposeful but should require user confirmation before execution.
Medium
Ruby/shell backtick execution
```bash
This is an executable shell example in the skill documentation. It is relevant to the workflow, but external commands should be reviewed before execution.
Medium
Ruby/shell backtick execution
```bash
This is an executable shell example in the skill documentation. It is relevant to the workflow, but external commands should be reviewed before execution.
Medium
Ruby/shell backtick execution
```bash
This is an executable shell example in the skill documentation. It is relevant to the workflow, but external commands should be reviewed before execution.
Medium
Hardcoded URL
When your agent-facing site is ready, add it to the ClawDirect directory at https://claw.direct so o
This documentation directs users or agents to the ClawDirect external service for directory operations. The network use is intentional, but it can affect account-owned listings and may involve payment.
Medium
Hardcoded URL
npx atxp-call https://claw.direct/mcp clawdirect_add '{
This documentation directs users or agents to the ClawDirect external service for directory operations. The network use is intentional, but it can affect account-owned listings and may involve payment.
Medium
Hardcoded URL
npx atxp-call https://claw.direct/mcp clawdirect_edit '{
This documentation directs users or agents to the ClawDirect external service for directory operations. The network use is intentional, but it can affect account-owned listings and may involve payment.
Medium
Hardcoded URL
npx atxp-call https://claw.direct/mcp clawdirect_delete '{
This documentation directs users or agents to the ClawDirect external service for directory operations. The network use is intentional, but it can affect account-owned listings and may involve payment.
Risk Factors
⚡ Contains scripts (1)
⚙️ External commands (68)
SKILL.md:122-127 SKILL.md:31 SKILL.md:35-62 SKILL.md:62-75 SKILL.md:75-80 SKILL.md:80-82 SKILL.md:82-83 SKILL.md:83-97 SKILL.md:97-99 SKILL.md:99-100 SKILL.md:100-103 SKILL.md:103-107 SKILL.md:107-109 SKILL.md:109-122 SKILL.md:128-137 SKILL.md:140-145 SKILL.md:147-150 SKILL.md:150-154 SKILL.md:154-156 SKILL.md:156-205 SKILL.md:205-209 SKILL.md:209-211 SKILL.md:211-224 SKILL.md:224-267 SKILL.md:267-271 SKILL.md:271-273 SKILL.md:273-324 SKILL.md:324-345 SKILL.md:345-346 SKILL.md:346-347 SKILL.md:347-352 SKILL.md:352-358 SKILL.md:358-361 SKILL.md:361-365 SKILL.md:365-377 SKILL.md:377-378 SKILL.md:378-384 SKILL.md:384 SKILL.md:386 SKILL.md:390 SKILL.md:392 SKILL.md:397-399 SKILL.md:399-400 SKILL.md:400-401 SKILL.md:401-407 SKILL.md:407-408 SKILL.md:408-411 SKILL.md:411-433 SKILL.md:433-434 SKILL.md:434-435 SKILL.md:435-436 SKILL.md:436-437 SKILL.md:437-447 SKILL.md:447-455 SKILL.md:455-460 SKILL.md:460-461 SKILL.md:461-462 SKILL.md:462-463 SKILL.md:463-464 SKILL.md:464 SKILL.md:470-475 SKILL.md:475-480 SKILL.md:480-481 SKILL.md:481-482 SKILL.md:482-483 SKILL.md:483-489 SKILL.md:489-493 SKILL.md:493-498
🌐 Network access (25)
SKILL.md:10 SKILL.md:20 SKILL.md:179 SKILL.md:324 SKILL.md:346 SKILL.md:347 SKILL.md:373 SKILL.md:378 SKILL.md:385 SKILL.md:391 SKILL.md:410 SKILL.md:417 SKILL.md:418 SKILL.md:419 SKILL.md:420 SKILL.md:421 SKILL.md:430 SKILL.md:439 SKILL.md:443 SKILL.md:448 SKILL.md:449 SKILL.md:471 SKILL.md:472 SKILL.md:490 SKILL.md:491
🔑 Env variables (10)
Audited by: codex View Audit History →
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/napoleond-clawdirect-dev/audits/4?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/napoleond-clawdirect-dev?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/napoleond-clawdirect-dev?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/napoleond-clawdirect-dev/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/napoleond-clawdirect-dev.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
napoleond. (2026). clawdirect-dev security audit report (audit version 4) [Author version unspecified]. Skillstore. https://skillstore.io/skills/napoleond-clawdirect-dev/audits/4BibTeX citation
@techreport{napoleond-napoleond-clawdirect-dev-2026,
author = {napoleond},
title = {clawdirect-dev security audit report (audit version 4)},
institution = {Skillstore},
year = {2026},
number = {4},
url = {https://skillstore.io/skills/napoleond-clawdirect-dev/audits/4},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "clawdirect-dev security audit report (audit version 4)"
version: "unspecified"
type: report
authors:
- name: "napoleond"
date-released: "2026-07-05"
url: "https://skillstore.io/skills/napoleond-clawdirect-dev/audits/4"
identifiers:
- type: other
value: "skillstore:napoleond-clawdirect-dev:audit:4"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: Medium 55
Architecture
85
Maintainability
87
Content
70
Community
78
Spec Compliance
What You Can Build
Build an Agent Login Flow
Create a browser authentication bridge for agents that already use MCP tools.
Add Paid MCP Actions
Design MCP tools that require ATXP authentication and optional payment before execution.
Publish an Agent Directory Entry
Prepare a ClawDirect listing so agents can discover and call a web service.
Try These Prompts
Plan My ATXP App
Use clawdirect-dev to outline an ATXP-enabled web app for my service. Include the MCP tools, browser flow, and storage choices.
Create Auth Middleware
Use clawdirect-dev to design cookie authentication for my Express API. Explain the files, routes, and validation steps I need.
Add a Paid Agent Action
Use clawdirect-dev to add a paid MCP action to my app. Include authentication, payment handling, and clear user-facing behavior.
Review Production Readiness
Use clawdirect-dev to review my planned ATXP browser authentication flow. Identify deployment, logging, cookie, and payment risks.
Best Practices
- Treat generated cookies as bearer credentials and protect them like secrets.
- Require explicit approval before running commands that install packages or spend money.
- Replace placeholders with project-specific names, domains, costs, and retention rules.
Avoid
- Do not use query strings for long-lived authentication tokens in production.
- Do not store raw bearer cookies without expiration, hashing, or revocation.
- Do not publish paid MCP tools without clear cost and confirmation behavior.
Frequently Asked Questions
What does this skill help build?
It helps build ATXP-enabled web apps that agents can use through browsers and MCP tools.
Does it require a specific backend stack?
Yes. The examples use Node.js, TypeScript, Express, SQLite, @longrun/turtle, and ATXP packages.
Can it add payments to MCP tools?
Yes. It shows a paid tool pattern using ATXP authentication and payment checks.
Is the cookie pattern production-ready?
No. The template needs stronger token handling, logging controls, expiration, and revocation before production use.
Does it create a full frontend?
No. It focuses on server, authentication, MCP, and publishing patterns for an agent-facing app.
Can Claude, Codex, and Claude Code use this skill?
Yes. The report marks the skill as compatible with Claude, Codex, and Claude Code.