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.
npx skills add Pawel-Kica/claude-code-100 --skill e2e-codexgit clone --depth 1 https://github.com/Pawel-Kica/claude-code-100Wrote 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.
[](https://agentmods.dev/skills/pawel-kica/claude-code-100/e2e-codex)<a href="https://agentmods.dev/skills/pawel-kica/claude-code-100/e2e-codex"><img src="https://agentmods.dev/badge/skills/pawel-kica/claude-code-100/e2e-codex.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00034 | $0.04188 |
| Opus 5 | $0.00017 | $0.02094 |
| Sonnet 5 | $0.00007 | $0.00838 |
| Haiku 4.5 | $0.00003 | $0.00419 |
Grade A, and why
e2e-codex 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 7d 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.
- No UI for the change, such as a scheduled task, API, or webhook: drive it anyway and say that is what happened. `curl` the real endpoint on the running server, or call the service function in the app's own shell. Asser How it starts
The opening of the file, as written. The whole thing — 238 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Give it a goal. Drive the real thing until you can answer whether it works. Fix what breaks, verify the fix, answer.
Scope
- Goal is whatever the user said, in prose.
- Args may be a spec, PRD, or plan path: read it for intent, the file is not the target.
- Nothing said: scope from
git statusplusgit diff --stat, or from the checklist written earlier in the conversation.
Wrong surface
The most common way this skill lies:
- Green on a page where the user is logged out, on the tab next to the right one, or on a modal that never opened.
- URL is the one intended, and the account, tenant, and role are the ones intended.
- Assert on state watched changing, then reload and assert it survived when persistence is part of the promise.
- What is asserted is the feature, not a lookalike such as editor source text, a hidden span, or the component's own progress label.
- Poll a count of real result nodes and a discriminating value. An SPA can keep the previous step mounted.
- The DOM can be right while the pixels are wrong. On canvas, WebGL, maps, charts, animation, and layout, inspect the screenshot.
Drive
Testing is always Chrome DevTools MCP. Playwright is for artifacts, nothing else.
Chrome DevTools MCP is the chrome-devtools server, always headless and isolated. It drives every verification run:
- Confirm it is loaded with
codex mcp list. When it is missing, add it once withcodex mcp add chrome-devtools -- npx -y chrome-devtools-mcp@latest --isolated --headless, then start a fresh Codex session so the tools appear. When it still does not load, report that this skill requires Chrome DevTools MCP instead of driving a different browser. - Click through by hand and read state back:
navigate_page,take_snapshotfor the accessibility tree,click,hover,fill,fill_form,type_text,press_key,select_page,wait_for,list_console_messages,list_network_requests,take_screenshot. No script to write. - A long or fiddly flow is not a reason to switch. Click it, do not script it.
- Interactions go through the interaction tools so hover, focus, and input events stay real.
evaluate_scriptreads state back and samples behaviour; it does not replace the click. - Each
take_snapshotmints fresh element uids. Re-snapshot after anything that mutates the DOM and act on the uid from that snapshot. - One shared browser instance serves every agent on this machine. Keep to your own page, select it explicitly before each payoff action, and expect a stolen tab whenever you do not.
- MCP file writes land where the server is permitted to write. A kept artifact whose destination sits outside that is Playwright's job.
What ships with it
3 files 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.
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.
- 7d ago First seen · 238 lines · 34 tokens per session scan A f8d76e2ee666
e2e-codex is a skill published in the GitHub repository Pawel-Kica/claude-code-100 (2 stars, last pushed 14d ago), licensed MIT. It adds 34 tokens to every session and 4,188 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 2026-08-31.
Other skills, from other repositories
browse
Drive a real browser through Aside: open a page, read it, click through a flow, take screenshots, check console errors. (gstack).
playwright-cli
A command-line tool for controlling Chromium, Firefox, and WebKit browsers, including navigation, page interaction, screenshots, PDFs, and recorded actions.
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
browser-qa
A browser-based quality check for deployed web pages and user flows. It uses browser automation to test rendering, navigation, forms, interactions, responsive behaviour, and accessibility-related issues.
test-electron-app
Drive the real running PostHog Electron app (live tRPC, workspace-server, real data) over CDP with agent-browser. Connect to the running app on port 9222, test desktop changes against a local Django stack, snapshot the accessibility tree, inspect network requests, and screenshot only when explicitly asked. Use when…
smoke-test
Run smoke tests against a deployed or local app based on your git diff. Each test uses Skyvern browser tools (navigate, act, validate, screenshot) with Chrome DevTools MCP as fallback. Posts screenshot evidence as PR comments.