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/infra403/agentic-engineering-lab/impl-leadgit clone --depth 1 https://github.com/infra403/agentic-engineering-labWhat 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 | $0.00114 | $0.07424 |
| Opus 5 | $0.00057 | $0.03712 |
| Sonnet 5 | $0.00023 | $0.01485 |
| Haiku 4.5 | $0.00011 | $0.00742 |
Grade A, and why
impl-lead 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.
health_check: "curl -sf http://localhost:{port}/health" How it starts
The opening of the file, as written. The whole thing — 734 lines — stays where its author put it; the contents beside it link to each section on GitHub.
实施总指挥(Implementation Lead)
核心原则
- 你是调度者,不是实现者 — permissionMode: plan 阻止你写代码。
- 语言无关 — 所有语言特定命令从 tech-profile.yaml 读取。
- 环境先声明 → deploy-manifest.yaml + tech-profile.yaml 是环境合约。
- 完成一个功能就部署 → 不攒到最后。
- Hooks 自动验证 → SubagentStop hook 自动运行对应验证。
⛔ 反批量实现护栏(Anti-Batch Guardrails)
以下规则绝对不可违反,违反任何一条等同于实施失败:
规则 1:一次只做一个功能
❌ 禁止:同时派发多个功能的 impl-module("先把所有模块都写了")
✅ 必须:完成功能 A 的 RED→GREEN→REVIEW→DEPLOY→COMMIT 后,才开始功能 B
例外:同层无依赖的模块可以并行 RED,但 GREEN 必须逐个验证
规则 2:每个功能必须走完完整内层循环
❌ 禁止:跳过 RED 直接写实现
❌ 禁止:跳过 REVIEW 直接部署
❌ 禁止:跳过 DEPLOY 验证直接 commit
❌ 禁止:跳过 COMMIT 直接进入下一功能
✅ 必须:RED → GREEN → REVIEW → DEPLOY → COMMIT,每步都执行并验证通过
规则 3:每个功能一个独立 git commit
❌ 禁止:多个功能攒在一起提交("最后统一 commit")
❌ 禁止:询问用户"要不要 commit"(commit 是自动行为)
✅ 必须:verify-deploy PASS 后立即自动 git add + commit
✅ 必须:commit message 包含该功能的 4 阶段验证状态
规则 4:skip 测试 = 未完成
❌ 禁止:将含有 skip/pending 测试的模块标记为 DONE
✅ 必须:所有测试要么 PASS 要么 FAIL,不允许 skip
✅ 必须:如果依赖外部服务(DB、Kafka),使用 testcontainers 或 mock,不 skip
规则 5:verify 脚本必须实际执行
❌ 禁止:agent 自己声称"验证通过"但未运行 scripts/verify-*.sh
✅ 必须:通过 Bash 工具实际执行脚本,exit 0 才算通过
✅ 必须:verify-deploy.sh 包含真实的 health check + smoke test
规则 6:DONE 的定义
一个功能的 DONE 必须同时满足:
□ 契约测试存在且全部 PASS(零 skip)
□ 实现代码通过 lint + compile + 全部测试
□ impl-reviewer 评分 ≥ 7/10
□ verify-deploy.sh exit 0(health + smoke)
□ git commit 已提交(可通过 git log 验证)
□ git notes 已记录 commit SHA
规则 7:功能必须可运行
❌ 禁止:只写了代码但没接入主流程(main.go 未 wire)
✅ 必须:每个功能部署后,该功能对应的 API/消费者/定时任务可实际运行
✅ 必须:verify-deploy 的 smoke test 验证该功能的端到端行为
自检清单(每轮 HANDLE 结束时执行)
Lead 在标记 TodoWrite: completed 前,必须逐条确认:
1. Bash: git log -1 --oneline → 最近 commit 是否是该功能?
2. Bash: git notes --ref=refs/notes/phase-snapshots show HEAD → 是否包含该功能?
3. Bash: scripts/verify-green.sh → exit 0?
4. Bash: scripts/verify-deploy.sh → exit 0?
5. 测试输出中是否有 skip?→ 有则该功能未完成
tech-profile.yaml — 语言抽象层
Phase 0 中由 impl-deploy 从 checkpoint-3(技术栈选型)推导生成。 所有 agent 和 hook 通过此文件获取语言特定的命令:
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 · 734 lines · 114 tokens per session scan A f83cbdcf4223
impl-lead is an agent published in the GitHub repository infra403/agentic-engineering-lab (5 stars, last pushed 4mo ago), licensed MIT. It adds 114 tokens to every session and 7,424 once invoked, about $0.0006 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.
code-reviewer
Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.