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/fastslack/mtw-e2e-runner/capturegit clone --depth 1 https://github.com/fastslack/mtw-e2e-runnerWhat 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 | $0.00009 | $0.00566 |
| Opus 5 | $0.00005 | $0.00283 |
| Sonnet 5 | $0.00002 | $0.00113 |
| Haiku 4.5 | $0.00001 | $0.00057 |
Grade A, and why
capture 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 2d 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.
- Do NOT try to manually authenticate, fetch tokens, write test files, or use curl. The tool handles auth automatically from project config. What it actually says
Quick Capture
Take a screenshot of any URL in one step. Handles pool checks and authentication automatically.
Workflow
-
Check pool — Call
e2e_pool_statusto confirm the Chrome pool is running. If not available, tell the user to runnpx e2e-runner pool startvia CLI and stop. -
Capture — Call
e2e_capturewith:url: The URL from the user's request (REQUIRED)cwd: The current working directory (REQUIRED — always pass this)fullPage: true if user says "full page", "full", "complete", or "toda la página"selector: CSS selector if user wants to wait for a specific elementdelay: milliseconds if user says "wait", "delay", or "espera"waitUntil: "domcontentloaded" if user mentions WebSocket, SSE, or real-time appsfilename: if user specifies a name
Authentication is automatic: the tool reads
authToken,authLoginEndpoint, andauthCredentialsfrom the project'se2e.config.js. You do NOT need to passauthTokenunless the user explicitly provides one. -
Show result — The tool returns the screenshot as an inline image. Show it to the user with the file path.
Arguments
The user passes the URL after the command:
/e2e-runner:capture http://localhost:3000/dashboard→ capture that URL/e2e-runner:capture http://localhost/concept-maps --full-page→ full page capture/e2e-runner:capture http://localhost/admin --delay 3000→ wait 3s before capture
If no URL is provided, ask the user for one.
Important
- Do NOT try to manually authenticate, fetch tokens, write test files, or use curl. The tool handles auth automatically from project config.
- Do NOT use the
e2e_runtool — this is a screenshot capture, not a test run. - Keep it simple: one
e2e_pool_statuscall + onee2e_capturecall. That's it.
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.
- 2d ago First seen · 46 lines · 9 tokens per session scan A 12b031b45916
capture is a command published in the GitHub repository fastslack/mtw-e2e-runner (3 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 9 tokens to every session and 566 once invoked, about $0.0000 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 commands, from other repositories
qa
Dispatch a verifiable browser task to the qa-tester subagent. Use for regression checks, smoke tests, and any task with a clean pass/fail outcome. Supports an EXHAUSTIVE mode that verifies every control behind an honesty gate.
interactive-dev
Start a persistent interactive development session with natural language browser control, auto-error monitoring, and full dev tool access.
verify-pr
Verify a PR's frontend changes through browser automation.
record
Record browser actions into a test definition.
playbook
Manage Mochi playbooks (list, show, run, delete). Plays through the browser MCP tools.
browser-stop
../../../commands/browser-stop.md.