tailtest-cline
tailtest-cline is a Cline plugin that automatically generates and runs tests every time the agent edits a source file. https://github.com/avansaber/tailtest-cline
README
tailtest-cline
You build. We cover. Across 8+ editors at once.
tailtest-cline is a Cline plugin that automatically generates and runs tests every time the agent edits a source file. Same R1-R15 rule layer as the other tailtest variants. Adversarial test mode (V13) shipped from day one.
Cline reach: VS Code, Cursor, JetBrains IDEs, Antigravity, Zed, Neovim, VSCodium, Windsurf, plus the Cline CLI. One plugin, eight-plus editors.
Full documentation at tailtest.com/docs/cline
What's different from the other variants
Cline does not have hooks (Claude Code's PostToolUse, Cursor's afterFileEdit, Codex's Stop). Instead, tailtest-cline uses three Cline-native primitives:
.clinerules/carries the rule layer (R1-R14 + R15 adversarial mode)tailtest-mcpserver (this repo) provides structured tools for the agent:tailtest_setup,tailtest_scenario_plan,tailtest_classify_failures,tailtest_pick_template,tailtest_ping- Memory Bank (
tailtestContext.md) holds the per-project profile across Cline sessions
The deterministic policy (depth tiers, R15 adversarial counts, framework templates, file paths) lives in MCP server code rather than rule text. The agent calls the tools when it needs scaffolding rather than relying on rule recall over long sessions.
Install
Three install paths:
1. Cline MCP Marketplace (recommended; one click)
Open Cline's Extensions panel inside your IDE. Search tailtest. Click install. Cline clones, installs deps, registers the server in cline_mcp_settings.json automatically.
2. Direct GitHub install
Paste https://github.com/avansaber/tailtest-cline into Cline's MCP server install dialog. Cline does the same clone + install + register flow.
3. Manual cline_mcp_settings.json edit
{
"mcpServers": {
"tailtest": {
"command": "python",
"args": ["-m", "src.server"],
"cwd": "/path/to/tailtest-cline/mcp_server",
"alwaysAllow": [
"tailtest_ping",
"tailtest_scenario_plan",
"tailtest_classify_failures",
"tailtest_pick_template"
]
}
}
}
First run
After install, type in Cline chat: "set up tailtest in this project".
Cline calls tailtest_setup. The tool detects language / framework / runner, writes .clinerules/, writes .clinerules/workflows/, seeds memory-bank/tailtestContext.md, and initialises .tailtest/config.json + .tailtest/session.json. Idempotent: existing files preserved.
Important: Cline does not auto-reload .clinerules mid-conversation. Start a new conversation (or reload the window) to activate.
Two operating modes
Manual mode (default)
You invoke the test cycle explicitly:
/tailtest-test <file>-- run the cycle on a specific file/tailtest hunt <file>-- one-shot adversarial pass on a specific file- Or natural language ("test the file I just edited")
Each step (file write, terminal command, MCP call) prompts for approval. Safer default for first-time users.
Auto mode (opt-in)
Enable Cline auto-approve for: Edit files (workspace), Execute safe commands, Use MCP servers. Then run /tailtest-mode auto and reload. After every edit, tailtest fires the test cycle automatically.
Plan / Act mode (Cline-native UX bonus)
Cline's Plan / Act toggle maps to tailtest's SCENARIO PLAN convention: Plan mode produces the SCENARIO PLAN (read-only); review and amend; Act mode writes the test file, runs it, and applies R12 classification. Recommended for high-stakes files.
Configuration
.tailtest/config.json:
{
"depth": "standard",
"mode": "manual"
}
Depth options:
simple-- 2-3 happy-path scenariosstandard-- 5-8 scenarios including 2+ adversarial probes (default)thorough-- 10-15 scenarios including 4+ adversarial probesadversarial-- 8-12 scenarios biased toward breakage
See tailtest.com/docs/config for all options and tailtest.com/docs/adversarial for adversarial mode details.
Other tailtest variants
Same R1-R15 rule layer, same adversarial test mode, different host integration. This repo is the Cline variant.
- tailtest -- Claude Code plugin (hook-driven)
- tailtest-cursor -- Cursor plugin (hook-driven)
- tailtest-codex -- Codex CLI plugin (hook-driven)
- tailtest-cline -- Cline plugin (MCP-driven; this repo)
See tailtest.com/demo/cline for a live walkthrough of this variant, or tailtest.com/comparison for a feature matrix across all four.
License
MIT.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。