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 agentmods add commands/alifanov/darkflow/check-uxgit clone --depth 1 https://github.com/alifanov/darkflowWrote 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/commands/alifanov/darkflow/check-ux)<a href="https://agentmods.dev/commands/alifanov/darkflow/check-ux"><img src="https://agentmods.dev/badge/commands/alifanov/darkflow/check-ux.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 | $0.00022 | $0.01415 |
| Opus 5 | $0.00011 | $0.00707 |
| Sonnet 5 | $0.00004 | $0.00283 |
| Haiku 4.5 | $0.00002 | $0.00142 |
Grade A, and why
check-ux 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 5d 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.
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 — 150 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The behavioural half of design review. check-design looks at surfaces; this one uses the
product — clicks through the real flows in a real browser, at a phone width and a desktop width,
and reports where a person would get stuck.
A screenshot of a page proves the page renders. It proves nothing about whether the third step of checkout can be completed. Only walking it does.
Proposal-only. Findings become tasks; nothing is changed here.
Step 1 — Read project config
Load the project config (contract in .darkflow.d/claude.md → Project config). Uses: language, stagingUrl, domain.
Walk stagingUrl when it is set, otherwise domain. Never walk a production flow that writes
real data — no live payments, no real emails, no destructive account actions. If only domain
is set, stop before any irreversible step and note in the snapshot which step you could not take.
If neither URL is set, there is nothing to walk: write nothing and stop. Do not substitute a
code reading — that is check-design's job, and a flow that was never walked must not be
reported as if it had been.
Step 2 — Pick the flows
Start from the data, not the sitemap. If analytics-review has run, it already named the
step that loses the most people:
cat .darkflow.d/state/metrics/analytics.json 2>/dev/null # visitors, window
rg -A30 '^## Analytics' docs/logs/*.md | tail -60 # the funnel table + headline drop
That step's flow is walked first, and its finding outranks anything found elsewhere — it is the one with a number attached. No such section (never ran, or a clean run) → pick flows as below.
Read docs/state/spec/flows/ when it exists — those are the flows the product says it has.
Otherwise derive them from sitemap.xml and the navigation: the entry path (landing → signup),
the core job the product exists for, and the money path (pricing → checkout) when there is one.
Three to five flows. More than that and nothing gets walked properly.
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.
- 5d ago First seen · 150 lines · 22 tokens per session scan A 4bee79735b16
check-ux is a command published in the GitHub repository alifanov/darkflow (2 stars, last pushed 6d ago), licensed MIT. It adds 22 tokens to every session and 1,415 once invoked, about $0.0001 per session on Opus 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 2026-08-31.
Other commands, from other repositories
record-workflow
Record browser interactions and generate a Playwright test with assertions and best-practice selectors.
e2e
Generate and run E2E tests with Playwright.
expect
Diff-aware AI browser testing — reads the git diff, maps changes to affected pages via the route map, generates a targeted test plan, and executes it via agent-browser (Rust daemon + CDP, ARIA-tree-first) with pass/fail reporting. Use when testing UI changes, verifying PRs before merge, or running regression checks on…
auto-verify
프론트엔드 UX 검증 — Playwright 기반 비주얼 검증을 실행합니다.
test-feature
Test a React Native feature on the running simulator/emulator. Verifies UI, user flows, and internal state. Generates a persistent Maestro test file.
record
Record a browser walkthrough of a URL using Antigravity (agy). Generates .webm video, screenshots, and a report. Auto-converts to MP4 if ffmpeg is available.