Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add timzaak/web-dev-skills/plugin install t-toolsWrote 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/timzaak/web-dev-skills/t-web-demo-run-all)<a href="https://agentmods.dev/skills/timzaak/web-dev-skills/t-web-demo-run-all"><img src="https://agentmods.dev/badge/skills/timzaak/web-dev-skills/t-web-demo-run-all/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/timzaak/web-dev-skills/t-web-demo-run-all"><img src="https://agentmods.dev/badge/skills/timzaak/web-dev-skills/t-web-demo-run-all.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 | $0.00094 | $0.03190 |
| Opus 5 | $0.00047 | $0.01595 |
| Sonnet 5 | $0.00019 | $0.00638 |
| Haiku 4.5 | $0.00009 | $0.00319 |
Grade A, and why
t-web-demo-run-all 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.
Chrome 扩展 Demo 的 fixture、--no-auto-env 参数传递、修复后重建和批次隔离按 ${CLAUDE_PLUGIN_ROOT}/protocols/web-demo-run-repair-contract.md 的“扩展 Demo 运行模式”；仅此模式下不执行默认 Web 环境 stop/start。
批量运行 Demo 测试（主会话逐文件驱动）
运行时边界：${CLAUDE_PLUGIN_ROOT}/protocols/runtime-boundaries.md（判断脚本入口或项目事实与插件默认冲突时读）
单文件运行与修复闭环：${CLAUDE_PLUGIN_ROOT}/protocols/web-demo-run-repair-contract.md（处理任一失败文件前读）
目标
- 自动发现除
live/、fixtures/、templates/、verification/和文件名含test-之外的全部 Demo 测试（发现规则脚本内固化，不在 skill 层重写）。 - 由当前主会话逐个文件驱动，每一步都是可观察、可中断的短 Bash / Agent 调用。
- 单文件失败时按
web-demo-run-repair-contract.md的修复闭环处理，每文件最多 6 次尝试。 - 文件之间重建 Demo 环境和数据容器，避免前一文件产生的业务数据影响后续测试。
- 每文件达到上限仍失败则标记
FAILED并继续下一个文件，不阻塞整批。 - 持续写盘批次状态，支持
continue从未完成文件的断点恢复；产出 Markdown + JSON 汇总报告。
两种模式
默认逐文件模式（无参数 / continue）：发现 → 逐文件跑 → 失败就当场修复 → 文件间数据隔离 → 收尾。每个文件独立诊断修复，互不参照。
扫描模式（scan）：先纯 Bash 预跑全部文件不修复，按归一化错误指纹聚类失败，再对每个 unique cluster 只修一次（用 representative 文件），最后回扫所有受影响文件。
何时用扫描模式：失败密集且疑似共享根因（如一批文件都挂在同一个失效 selector 或同一个后端接口变更上）。先用 ${CLAUDE_PLUGIN_ROOT}/guides/web-demo/batch-token-profiling.md 的决策树判断；孤立失败为主或文件数少（< 5）时用默认模式，扫描模式的聚类开销不划算。不确定时先跑默认模式一次，再用该指南读画像决定。
执行流程（默认逐文件模式）
A. 发现阶段（一次性，脚本辅助）
uv run scripts/web-demo-run-all.py discover [--filter-file <path>] # fresh
uv run scripts/web-demo-run-all.py discover continue # 断点恢复
stdout 打印单行 JSON，必须解析它，后续循环使用其中的 discovered_files（仓库相对路径列表）、batch_run_id、json_report（批次产物路径）和 resume_index（fresh 为 0）等字段，字段语义以脚本输出为准。脚本同时已写好初始批次 JSON（fresh）或已截断/重置好状态（continue）。脚本返回非零时按其错误信息处理（如 continue 找不到可继续内容则终止）。
B. 主循环：对 discovered_files[resume_index:] 逐个文件
对每个文件记其 zero_based_index（在 discovered_files 中的全量下标，不是切片下标），依次执行：
B1. 写断点
uv run scripts/web-demo-run-all.py checkpoint --json <json_report> --index <zero_based_index>
B2. 运行整个测试文件
uv run scripts/web-demo-test-runner.py "<rel_path>" --run-id "<batch_run_id>-<file_key>-initial" --mode fast
web-demo-test-runner.py 会自检并按需启动/恢复 demo 环境。解析其最后一行 Result: {...}，按 web-demo-run-repair-contract.md 的 Runner 输出适配映射为 snake_case 批次 entry 字段。file_key 必须由相对路径的稳定 slug 或短哈希生成，避免不同目录的同名文件冲突。
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 Changed · +3 lines 318abebef1d7
- 6d ago Changed · -84 lines 7132c21abfed
- 11d ago First seen · 266 lines · 94 tokens per session scan A 60d76ecae88b
t-web-demo-run-all is a skill published in the GitHub repository timzaak/web-dev-skills (74 stars, last pushed today), licensed Apache-2.0. It adds 94 tokens to every session and 3,190 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
azure-playwright-workspaces
Expert knowledge for Playwright Workspaces development including troubleshooting, best practices, decision making, limits & quotas, security, and configuration. Use when managing Playwright Testing workspaces, tokens/RBAC, quotas, monitoring/metrics, or run/AADSTS7000112 issues, and other Playwright Workspaces related…
verify
Prove a change works end to end through acceptance criteria, focused project gates, real user or caller flows, edge cases, and reliability checks. Use before a PR, when the user asks whether a change actually works, or as the first stage of $ship even when automated tests are green.
integration-testing
Guidance for writing integration and end-to-end tests that use real databases, HTTP services, or message queues. It also covers keeping each test independent and repeatable.
e2e-testing
Guide for running end-to-end tests of the Qwen Code CLI, including headless mode, MCP server testing, and API traffic inspection. Use this skill whenever you need to verify CLI behavior with real model calls, reproduce user-reported bugs end-to-end, test MCP tool integrations, or inspect raw API request/response…
terminal-capture
Automates terminal UI screenshot testing for CLI commands. Applies when reviewing PRs that affect CLI output, testing slash commands (/about, /context, /auth, /export), generating visual documentation, or when 'terminal screenshot', 'CLI test', 'visual test', or 'terminal-capture' is mentioned.
agent-reproduce-align
Use after a Codex or Claude Code feature has been implemented in Qwen Code to run the selected reference agent and Qwen Code under the same scenario, capture HTTP and terminal traces, compare request bodies, tool/function schemas, outputs, and iterate until the reproduced behavior is close enough.