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 skills/coffeecheese/easy-prd-testing/execution-gatenpx skills add CoffeeCheese/easy-prd-testing --skill execution-gategit clone --depth 1 https://github.com/CoffeeCheese/easy-prd-testingWhat 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 | $0.00034 | $0.01237 |
| Opus 5 | $0.00017 | $0.00619 |
| Sonnet 5 | $0.00007 | $0.00247 |
| Haiku 4.5 | $0.00003 | $0.00124 |
Grade A, and why
execution-gate 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 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.
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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Execution Gate
Goal
Confirm every prerequisite needed before page execution, including target and service lifecycle ownership.
Required Files
The module artifact directory must contain readable:
01-模块拆解.md02-测试用例.md03-执行清单.md04-测试数据与账号.md
The conversation or visible task state must show explicit user confirmation of 01-04. Invocation after planning is not enough. If explicit confirmation is absent, stop and ask the user to confirm 01-04 before collecting execution prerequisites.
Check file existence separately from confirmation.
Confirmation Order
Resolve prerequisites in this order. Before asking, use the input path or URL, read-only port or reachability probes, the existing execution plan, and 01-04 to infer what can be recorded safely. Ask only for the earliest item that still requires user information or authorization:
- Module artifact directory, if the
01-04files cannot be located. - Explicit user confirmation of
01-04, if not observable in conversation or visible task state. - Test environment address and module test path, only if they cannot be derived reliably.
- Target preflight, using the rules below. Do not ask the user to confirm target type, service state, safe ownership defaults, readiness basis, or cleanup responsibility when they can be derived reliably. Ask only when:
- A stopped local service must be started for this run: collect its startup command and responsibility, then request one explicit startup authorization.
- Resource ownership cannot be determined safely and would affect start, stop, or cleanup behavior.
- No observable page-readiness basis can be derived from the test cases, target, or available page information.
- Login mode: specified account, reuse current session, or no login.
- For specified account: account, password retrieval method, and permission scope.
- Execution mode: single agent or priority-based multi-agent.
- Execution backend policy: default
agent-browser, diagnostic Chrome DevTools CLI, final fallback Chrome DevTools MCP. - Runtime dependency check: run
scripts/check-deps.shwhen available, or manually check the same dependency layers when this skill is installed without repository scripts. - Playwright Test policy: generate regression script drafts only when the user requests script persistence or when P0/P1 failures need future regression.
- High-risk action authorization: allow real submission, or verify only entry, display, validation, secondary confirmation, and messages.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 89 lines · 34 tokens per session scan A f97e898effea
execution-gate is a skill published in the GitHub repository CoffeeCheese/easy-prd-testing (2 stars, last pushed 20d ago), licensed MIT. It adds 34 tokens to every session and 1,237 once invoked, about $0.0002 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 skills, from other repositories
playwright-cli
Automate browser interactions, test web pages and work with Playwright tests.
playwright-dev
Explains how to develop Playwright - add APIs, MCP tools, CLI commands, and vendor dependencies.
playwright-recording
Record browser interactions as video using Playwright. Use for capturing demo videos, app walkthroughs, and UI flows for Remotion videos. Triggers include recording a demo, capturing browser video, screen recording a website, or creating walkthrough footage.
webcmd-browser
Use when a live browser task requires Playwright interaction, authenticated handoff, visible UI verification, or ad-hoc page inspection.
browser-automation
Playwright-based browser automation patterns for autonomous web interaction.
moli-webfetch
Fetch, inspect, crawl, and capture live, JavaScript-rendered websites with Moli. Use when Codex needs current web content, web research, fact lookup, link following, a bounded crawl, client-rendered or response-gated content, network diagnostics, or a standalone HTML, Markdown, JSON, semantic-tree, viewport or…