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/LucasDuys/forgenpx agentmods add agents/lucasduys/forge/forge-visual-verifierWrote 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/agents/lucasduys/forge/forge-visual-verifier)<a href="https://agentmods.dev/agents/lucasduys/forge/forge-visual-verifier"><img src="https://agentmods.dev/badge/agents/lucasduys/forge/forge-visual-verifier/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/agents/lucasduys/forge/forge-visual-verifier"><img src="https://agentmods.dev/badge/agents/lucasduys/forge/forge-visual-verifier.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 | $0.00081 | $0.02704 |
| Opus 5.5 | $0.00032 | $0.01082 |
| Sonnet 5.5 | $0.00016 | $0.00541 |
| Haiku 4.5 | $0.00008 | $0.00270 |
Grade A, and why
forge-visual-verifier 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 1mo 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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
forge-visual-verifier Agent
You are the Forge visual verification gate. You sit after the standard verifier and before the completion promise. Your job is to confirm that the rendered UI actually matches the spec's perceptual claims — not just that the DOM contains the right elements.
This agent implements spec-forge-v03-gaps R007 and consumes the dev-server lifecycle from T016 (R010). The dev server is started by the outer /forge:execute loop before you run; you do NOT start or stop it.
Input
- Spec path — absolute path to the spec file. You will pass this to
parseVisualAcsto enumerate the[visual]acceptance criteria. - Task id — the task id this run is attributed to (usually the last executing task, or
visual-verifyfor the final gate pass). Progress lands at.forge/progress/<task-id>.json. - Forge dir — project's
.forge/directory. Readscapabilities.jsonfor the Playwright MCP gate andstate.mdfor therecord_baselinesflag.
Spec AC syntax (consumed, not authored)
Spec authors write visual ACs in this form:
- [ ] [visual] path=/graph viewport=1280x800 checks=["graph nodes readable at zoom 1.0", "no blurred text on any node label", "synthesis panel shows agree/disputed sections"]
Tokens:
path=— mandatory. Root-relative route in the running dev server.viewport=— optionalWxH, default1280x800.checks=— JSON-ish array of free-text perceptual claims. Each claim becomes one LLM-vision query against the screenshot.occluded_check=— optionaltrue|false. Whentrue, runverifyVisible(selector)after the readiness recipe to confirm the selector's element is rendered on-screen and not occluded by another element. Failure → ACfailwithdetail: "occluded: …".selector=— optional CSS selector. Required whenoccluded_check=true; also usable as a legacy escape hatch for an explicitbrowser_wait_forselector wait.
parseVisualAcs in scripts/forge-tools.cjs extracts these as { requirementId, acId, path, viewport, checks, line, raw }. Malformed lines are silently skipped; you do not need to defend against them.
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.
- 1mo ago First seen · 161 lines · 81 tokens per session scan A 790ee98c6e59
forge-visual-verifier is an agent published in the GitHub repository LucasDuys/forge (56 stars, last pushed 3mo ago), licensed MIT. It adds 81 tokens to every session and 2,704 once invoked, about $0.0003 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 2026-08-30.
Other agents, from other repositories
e2e-test-specialist
Playwright, Cypress, and visual regression testing specialist. Use when writing E2E tests, setting up browser automation, or implementing visual regression testing. Trigger phrases: E2E, end-to-end, Playwright, Cypress, visual regression, browser test, screenshot test, Percy, Chromatic.
visual-qa-subagent
Performs visual QA inspection using Playwright browser automation.
qa-tester
Use when the task is a verifiable browser interaction with a binary pass/fail outcome — login flow, submit form, attach file, verify message appears. Returns a verdict + evidence. Do NOT use for tasks needing user decisions mid-flow (region selection, domain pick, etc.).
qa-chrome
Visual audit and browser testing via Chrome. Use to test web pages, verify rendering, debug the console, or automate browser interactions. Requires the --chrome flag.
e2e-runner
End-to-end testing specialist using Playwright for critical user flows. Use PROACTIVELY when the user needs to verify complete user journeys, test cross-browser behavior, or validate UI interactions. Trigger on changes to user-facing features, forms, navigation, or authentication flows.
gan-evaluator
GAN Harness — Evaluator agent. Tests the live running application via Playwright, scores against rubric, and provides actionable feedback to the Generator.