mcp-csp-investigation

mcp-csp-investigation is a skill for Claude Code from a5c-ai/babysitter. It costs 43 tokens per session (2,247 once invoked), scanned A, original, MIT.

A Content Security Policy audit for MCP Apps, which are small apps shown inside sandboxed frames. It finds the outside web addresses used by the app and produces the declarations needed to allow them.

In plain words
What is it for?
Use it to inspect built files, trace each network address back to source code, and configure the policy passed to the app resource.
Why use it?
Without the right policy, requests for APIs, scripts, fonts, images, or WebSockets can fail silently, making the app appear broken without a clear error.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

Good fit Use it to inspect built files, trace each network address back to source code, and configure the policy passed to the app resource.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/a5c-ai/babysitter/mcp-csp-investigation
About the project

Babysitter is a workflow engine for AI coding agents that enforces predefined steps, quality checks, human approvals, and decision records. It is used to coordinate complex, repeatable agent workflows across supported coding tools. The catalogue contains skills, agents, instructions, settings, a plugin, and an MCP integration for its workflow.

a5c-ai/babysitter · 1,788 stars · on GitHub · a5c.ai

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add a5c-ai/babysitter --skill mcp-csp-investigation
Clone the repo
git clone --depth 1 https://github.com/a5c-ai/babysitter

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for mcp-csp-investigation

README.md
[![agentmods](https://agentmods.dev/badge/skills/a5c-ai/babysitter/mcp-csp-investigation/github.svg)](https://agentmods.dev/skills/a5c-ai/babysitter/mcp-csp-investigation)
Your own site
<a href="https://agentmods.dev/skills/a5c-ai/babysitter/mcp-csp-investigation"><img src="https://agentmods.dev/badge/skills/a5c-ai/babysitter/mcp-csp-investigation/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for mcp-csp-investigation

Your own site · 80×15
<a href="https://agentmods.dev/skills/a5c-ai/babysitter/mcp-csp-investigation"><img src="https://agentmods.dev/badge/skills/a5c-ai/babysitter/mcp-csp-investigation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 43 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,247 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to high

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • high Anti-Refusal · line 22
    Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.
    Fix: Remove instructions that suppress warnings, disclaimers, or ethical commentary. Let the agent surface safety-relevant caveats to the user.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00043 $0.02247
Opus 5 $0.00022 $0.01123
Sonnet 5 $0.00009 $0.00449
Haiku 4.5 $0.00004 $0.00225

Measured 6d ago against content hash 70e54f99815f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

mcp-csp-investigation scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

fetch(apiUrl); // Works for default, fails for custom API_URL
library/specializations/ai-agents-conversational/skills/mcp-csp-investigation/SKILL.md · 259 lines

How it starts

The opening of the file, as written. The whole thing — 259 lines — stays where its author put it; the contents beside it link to each section on GitHub.

mcp-csp-investigation

Perform exhaustive Content Security Policy audits for MCP Apps running in sandboxed iframes where all network requests fail SILENTLY without proper CSP declarations.

Overview

MCP Apps run in sandboxed iframes with no same-origin server. This means:

  • ALL network requests fail silently without CSP -- no errors, no warnings, just silent failure
  • Every external origin (CDN, API, font, image, WebSocket) must be declared in CSP
  • CSP is configured in the contents[] return from the registerAppResource read callback
  • Missing even ONE origin causes that resource to silently not load

This skill provides a systematic methodology for discovering every network origin an app uses, tracing each to its source, and generating the correct CSP configuration.

Capabilities

Build Output Analysis

  • Build the application and capture all output files (HTML, CSS, JS, assets)
  • Search every file for network origin references
  • Identify fetch/XHR targets, script sources, link hrefs, image sources, font URLs, iframe sources, WebSocket endpoints

Origin Tracing

  • Trace each discovered origin to its source in the codebase
  • Classify as: hardcoded constant, environment variable, or conditional logic
  • Document whether origin is universal, dev-only, or prod-only
  • Check third-party libraries for hidden network requests (analytics, telemetry, CDN fallbacks)

CSP Domain Categorization

  • resourceDomains: Scripts, stylesheets, images, fonts (maps to script-src, style-src, img-src, font-src)
  • connectDomains: fetch/XHR targets, WebSocket endpoints (maps to connect-src)
  • frameDomains: Nested iframes (maps to frame-src)

CSP Configuration Generation

  • Generate the CSP object for registerAppResource read callback
  • Handle environment-specific origins with proper conditional logic
  • Verify conditional origins have matching runtime URL and CSP entry

Usage

Step 1: Build the Application

# Build to produce final output files
npm run build

# Identify all output files
find dist/ -type f \( -name "*.html" -o -name "*.js" -o -name "*.css" \)

Read the full file on GitHub · 259 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 259 lines · 43 tokens per session scan A 70e54f99815f

Subscribe to this mod's changes

mcp-csp-investigation is a skill published in the GitHub repository a5c-ai/babysitter (1,788 stars, last pushed 6d ago), licensed MIT. It adds 43 tokens to every session and 2,247 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-05.