Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/kwannoel/the-controllernpx agentmods add skills/kwannoel/the-controller/the-controller-e2e-evalWrote 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/kwannoel/the-controller/the-controller-e2e-eval)<a href="https://agentmods.dev/skills/kwannoel/the-controller/the-controller-e2e-eval"><img src="https://agentmods.dev/badge/skills/kwannoel/the-controller/the-controller-e2e-eval/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.
<a href="https://agentmods.dev/skills/kwannoel/the-controller/the-controller-e2e-eval"><img src="https://agentmods.dev/badge/skills/kwannoel/the-controller/the-controller-e2e-eval.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00039 | $0.01723 |
| Opus 5.5 | $0.00016 | $0.00689 |
| Sonnet 5 | $0.00008 | $0.00345 |
| Haiku 4.5 | $0.00004 | $0.00172 |
Grade A, and why
the-controller-e2e-eval scanned grade A with 0 findings 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 yesterday.
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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.
E2E Eval — User Story Validation
When to Use
You've made a UI change and need to prove it works from the user's perspective — not "it compiles" or "the DOM element exists," but "the user can do the thing and see the result."
Hard Rules
These are non-negotiable. Every test must follow all of them.
1. User story first, always
Before writing a single line of test code, state the user story:
- Actor: Who is performing the action (e.g., "a user in Architecture mode")
- Action: What they physically do (e.g., "presses
r") - Outcome: What they expect to see (e.g., "an architecture diagram renders in the diagram pane")
The test's final assertion must verify the outcome. If the outcome doesn't happen, the test fails. No exceptions.
2. No silent passes
BANNED — the if (visible) escape hatch:
// NEVER DO THIS — silently passes when the feature is broken
const visible = await foo.isVisible().catch(() => false);
if (visible) {
await expect(foo).toHaveText("expected");
}
REQUIRED — unconditional assertion with a timeout:
// DO THIS — fails loudly when the feature is broken
await expect(foo).toBeVisible({ timeout: 10_000 });
await expect(foo).toHaveText("expected");
If the element should be there, assert it. If it might legitimately not be there, that's a different test or a precondition you need to set up.
3. No mocking the interface
- No
page.evaluate()to set store state when a real interaction exists. If the user presses Space to open the workspace picker, the test presses Space. - No synthetic
dispatchEvent(new KeyboardEvent(...))— usepage.keyboard.press(). - No stubbing API responses — the test runs against real servers via
eval.sh.
The only acceptable use of page.evaluate() is to read state for assertions (e.g., checking a computed style), never to write state that bypasses the UI.
4. Assert outcomes, not implementation
WRONG — asserting CSS values and class names:
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.
- yesterday First seen · 185 lines · 39 tokens per session scan A 165c45f59b76
the-controller-e2e-eval is a skill published in the GitHub repository kwannoel/the-controller (13 stars, last pushed 4mo ago), licensed MIT. It adds 39 tokens to every session and 1,723 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-23.
Other skills, from other repositories
eval-dataset-design
A guide to creating reliable test sets for AI agents. Each evaluation task includes the request, the simulated user's behaviour, a reset starting state, and a success check that can be independently verified.
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
test-warp-ui
Guides testing Warp UI features and changes using the computer use tool. Use this skill only when computer-use testing was requested (explicit request or accepted offer) and the computeruse tool is available to the agent. Covers launching Warp and verifying UI behavior.
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…
pyats-dynamic-test
Generate and execute deterministic pyATS aetest validation scripts - interface state, OSPF neighbors, BGP paths, ping matrices, and custom compliance tests. Use when writing a network test, validating post-change state, running pass/fail checks, or building automated regression tests. For Arista EOS devices, prefer…