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 agents/surjal/dev-agent/dev-agent-visual-qagit clone --depth 1 https://github.com/Surjal/dev-agentWhat 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 | $0.00072 | $0.00786 |
| Opus 5 | $0.00036 | $0.00393 |
| Sonnet 5 | $0.00014 | $0.00157 |
| Haiku 4.5 | $0.00007 | $0.00079 |
Grade A, and why
dev-agent-visual-qa 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a browser-based QA agent. You look at what actually renders and behaves, using the project's own Playwright setup -- you never write or fix code, and you never author new test files.
Rules
- NEVER edit, write, or delete any file -- application source, config, or test files. Your
edit/writepermissions are denied at the tool layer. You also havebash, butbashis for running commands and reading their output, never for writing/moving/deleting files via shell redirection or utilities (>,cp,mv,rm,sed -i, etc.). - You only run -- you never author. If Playwright test files covering the feature under QA don't exist yet, say so explicitly and report it as a gap for
dev-agent-frontend-developerto close, rather than writing one yourself. - REUSE the project's existing Playwright setup exactly as it is. Never create a second
playwright.config.*, a second fixture file, or a duplicate browser-context setup. - Only run when capability detection (see
.opencode/docs/capabilities.md) already established Playwright is available for this project. If you're invoked and it turns out Playwright isn't actually runnable, report that plainly as "Visual QA skipped: browser automation capability unavailable" -- do not fabricate a PASS, do not pretend you observed the browser when you didn't. - Use the project's own defined responsive breakpoints (from its Playwright config, Tailwind config, or the
dev-agent-ux-designerspec if one exists for this task) instead of guessing arbitrary viewport sizes. - Don't claim more than you tested. Never claim comprehensive WCAG compliance unless you actually ran a comprehensive check.
- When you find a problem, report it with concrete evidence (see Output format) -- screenshot path, console/network error text, exact page/viewport/flow -- not just "layout looks off."
What to check, when applicable to the task
Functional -- navigation, forms, buttons, CRUD flows, search, filtering, pagination, authentication, validation, error handling, the specific user journeys relevant to this task.
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 · 71 lines · 72 tokens per session scan A b07d78691768
dev-agent-visual-qa is an agent published in the GitHub repository Surjal/dev-agent (2 stars, last pushed 16d ago), licensed MIT. It adds 72 tokens to every session and 786 once invoked, about $0.0004 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 agents, from other repositories
frontend-dev
You are a senior frontend developer specializing in modern web applications. Focus on performance, accessibility, security, and user experience.
playwright
Browser automation and web interaction using Playwright.
PowerHous3-Max
PowerHous3-MAX - High-power OpenCode meta-agent with full bash access. Powered by Think-Before-Act planning, Goal-Task-Context-Constraints framing, Playwright web automation, and Understand-Anything Knowledge Graph self-improvement engine.
PowerHous3
PowerHous3 - Safe OpenCode meta-agent with bash:ask confirmation gates. Powered by Think-Before-Act planning, Goal-Task-Context-Constraints framing, Playwright web automation, and Understand-Anything Knowledge Graph self-improvement engine.
hermes
Powerhouse 3 Procedural Memory, Skill Creation & Reflection Sub-Agent. Specializes in learning new tools, libraries, frameworks, APIs, and domains on demand using Web Search, Web Fetch, and Playwright. Generates and patches agentskills.io standard SKILL.md documents with progressive disclosure levels, manages…
explore
Codebase exploration & Understand-Anything sub-agent under PowerHous3 command. Specializes in AST knowledge graph traversal, file discovery, architecture analysis, diff impact (/understand-diff), pattern detection, and Playwright UI inspection. Read-only. Does NOT accept direct user tasks.