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 skills add echoVic/blade-code --skill real-api-qualification-and-e2egit clone --depth 1 https://github.com/echoVic/blade-codeWrote 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/echovic/blade-code/real-api-qualification-and-e2e)<a href="https://agentmods.dev/skills/echovic/blade-code/real-api-qualification-and-e2e"><img src="https://agentmods.dev/badge/skills/echovic/blade-code/real-api-qualification-and-e2e.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00182 | $0.02718 |
| Opus 5 | $0.00091 | $0.01359 |
| Sonnet 5 | $0.00036 | $0.00544 |
| Haiku 4.5 | $0.00018 | $0.00272 |
Grade A, and why
knowledge-engineering-quality-and-delivery-real-api-qualification-and-e2e 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 4d 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Module Structure
真实 API 层以生产 Agent/协议入口为被测对象，用显式模型×surface 矩阵、宿主可验证副作用和严格资源回收把“模型说完成了”提升为可发布证据。
Directory Layout
packages/cli/tests/integration/real-api/— 真实 Provider 能力、恢复、故障注入和跨表面轨迹packages/cli/tests/support/— Headless、raw PTY、production Web、ACP 驱动及透明代理packages/cli/tests/e2e/— 本地模拟 E2E；不承担付费发布资格packages/cli/scripts/qualification.ts、packages/cli/scripts/qualify.ts— 本地/生产门禁计划与顺序执行packages/cli/scripts/real-api-credentials.ts— 受限凭据文件解析和环境投影packages/cli/scripts/test-config.js—realApi与 38 个发布阻断 trajectory 的显式配置docs/testing/、docs/en/testing/— 双语资格契约和冻结候选证据
Key Entry Points
realApiQualificationinpackages/cli/scripts/test-config.js— 发布阻断文件白名单、90 分钟 watchdog 和 release 环境resolveRequiredDeepSeekQualificationModels()inpackages/cli/tests/integration/real-api/testConfig.ts— 强制解析 Flash/ProbuildRealApiRuntimeConfig()inpackages/cli/tests/integration/real-api/testConfig.ts— 为每个测试渠道生成隔离的生产 Runtime 配置packages/cli/tests/support/setup.real-api.ts— 初始化真实模型并避开普通测试 mockmaterializeRealApiEnvironment()inpackages/cli/scripts/real-api-credentials.ts— 合并显式环境与安全凭据文件startRecordingProviderProxy()inpackages/cli/tests/support/recordingProviderProxy.ts— 记录、阻塞和转发真实 Provider 请求
Branching Table
| 维度 | 分支 A | 分支 B |
|---|---|---|
| 启用状态 | 未设置 REAL_API_TEST=1 时大多数 trajectory 跳过 |
启用后若没有任何可用凭据则 setup/import fail closed |
| 执行集合 | test:real-api 运行完整 soak 集合，watchdog 60 分钟 |
test:real-api:qualification 只运行显式 38 文件白名单，watchdog 90 分钟 |
| 框架重试 | 普通 real-api project 允许一次 Vitest retry | REAL_API_RELEASE_MATRIX=1 时 retry 固定为 0 |
| 必需模型 | 普通 soak 使用当前已配置模型 | 发布资格强制 DeepSeek deepseek-v4-flash 与 deepseek-v4-pro |
| 可选 Provider | 默认排除 domestic 凭据并保留必需渠道 | REAL_API_INCLUDE_OPTIONAL_PROVIDERS=1 才把 domestic 加入付费环境 |
| surface 集合 | 某些高成本轨迹在 release helper 下移除 raw PTY，形成 6 格 | Browser、token-budget、large-prompt 等关键轨迹固定四 surface，形成 8 格 |
| 浏览器状态 | Chromium preflight 失败时顺序门禁立即停止 | preflight 成功后才进入付费 trajectory |
| 凭据来源 | 显式 Provider 环境变量形成调用方选择的渠道集合 | 无显式 key 时可从受限凭据文件或当前 Blade 凭据解析 |
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.
- 4d ago First seen · 91 lines · 182 tokens per session scan A 1b88d4de76da
knowledge-engineering-quality-and-delivery-real-api-qualification-and-e2e is a skill published in the GitHub repository echoVic/blade-code (178 stars, last pushed 2d ago), licensed MIT. It adds 182 tokens to every session and 2,718 once invoked, about $0.0009 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 2026-09-03.
Other skills, from other repositories
dogfood
Exploratory QA of web apps: find bugs, evidence, reports.
langbot-testing
Test LangBot WebUI and core product flows with an automated browser and backend logs. Use when validating the configured LangBot frontend, pipeline Debug Chat, model provider setup and test buttons, bot and knowledge-base UI flows, or troubleshooting failed LangBot end-to-end tests.
baguette
Drive iOS simulators programmatically via the baguette CLI — taps, swipes, multi-finger gestures, hardware buttons (Home / Lock / Volume / Action / Power), ASCII keyboard text, and frame capture, all without opening Xcode. Use when: (1) an agent needs to drive a booted iOS simulator from a script — tap a coordinate…
local-test
Build, run, and test IronClaw locally using Docker containers and Chrome MCP browser automation.
app_verification
Verify a newly created or newly installed Rome app after coding:appcreation or coding:workflowcreation finishes installing it. Use only as a second-pass verifier: receive the app id, source root, expected behavior, and local dashboard/API base URL from the creator, then visit the installed app, exercise its safe…
tmux-manual-qa
Run a single manual tmux-based QA scenario for the todo continuation feature against the real CLI (./pi-test.sh) in an interactive TUI. Captures scrollback, asserts deterministic pass/fail count markers, and cleans up test fixtures. Use only for the manual-qa milestone features.