open-pr: Skill for Claude Code

.claude/skills/e2e-loop/SKILL.md

e2e-loop is a skill for Claude Code, Cursor from TOMOSIA-VIETNAM/open-pr. It costs 85 tokens per session (1,396 once invoked), scanned A, original, MIT.

A repeated test-and-review process for changes to the open-pr source code. It runs a real review, has a separate agent grade the result against a checklist, fixes identified problems, and repeats for a limited number of rounds.

In plain words
What is it for?
Use it after modifying files under src/ when you need end-to-end evidence that reviews still pass the project’s checklist.
Why use it?
It tests whether prompt changes produce the intended review behavior, which ordinary automated tests cannot fully prove.

Skill for Claude CodeCursor

Written for Claude Code and Cursor: ${CLAUDE_PLUGIN_ROOT} variable, but also shipped in a Cursor plugin. Also seen: mentions CLAUDE.md; mentions subagents.

This is TOMOSIA-VIETNAM/open-pr's own configuration. It tells Claude Code and Cursor how to work on open-pr itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything open-pr configures →

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the open-pr plugin — 6 skills, 1 command, 1 plugin shipped together

Reuse

Borrowing it

Nothing to install: this file belongs to TOMOSIA-VIETNAM/open-pr. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/TOMOSIA-VIETNAM/open-pr/main/.claude/skills/e2e-loop/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/TOMOSIA-VIETNAM/open-pr

Made for: Claude Code, Cursor.

Or install open-pr, the plugin that ships this one along with the rest of its 6 skills, 1 command, 1 plugin.

Wrote 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.

agentmods badge for e2e-loop

README.md
[![agentmods](https://agentmods.dev/badge/skills/tomosia-vietnam/open-pr/e2e-loop.svg)](https://agentmods.dev/skills/tomosia-vietnam/open-pr/e2e-loop)
Your own site
<a href="https://agentmods.dev/skills/tomosia-vietnam/open-pr/e2e-loop"><img src="https://agentmods.dev/badge/skills/tomosia-vietnam/open-pr/e2e-loop.svg" alt="Measured on agentmods" height="20"></a>
Per session 85 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,396 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00085 $0.01396
Opus 5 $0.00043 $0.00698
Sonnet 5 $0.00017 $0.00279
Haiku 4.5 $0.00009 $0.00140

Measured 8d ago against content hash 4963ab7b0863, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

e2e-loop 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 8d 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.

.claude/skills/e2e-loop/SKILL.md · 113 lines

How it starts

The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.

e2e loop

The unit suite proves the prompt graph is well-formed. It cannot prove a review still comes out right. This closes that gap: run → grade → diagnose → fix → re-run, with the running and the grading done by DIFFERENT agents so neither marks its own work.

Costs real money and posts to a real PR on open-pr-test. Never start a round the user did not ask for.

Why subagents rather than this session

A dev session already knows which defects the fixture plants and which rule was just edited. Reviewing from that context tests the session's memory, not the prompts. Each round therefore spawns fresh agents that were told nothing beyond what a real user's session gets.

Round budget

Ask for max_rounds if the user did not say; default 2. Stop early when every checklist row passes, or when a round produces no NEW passing row — a loop that keeps editing without moving the score is the failure mode this skill is most likely to hit.

Preflight

  1. scripts/check.sh <base-ref> must be green. A red suite makes every later verdict unreadable.
  2. e2e/bootstrap.sh --pr <n> [--vendor …] if no fixture PR is open for this round.
  3. python3 scripts/vendor_lint.py --pr <n> — every documented Fetch command must run. A broken vendor command wastes a whole round: the review fails at fetch and every checklist row reads fail for a reason that has nothing to do with the rules being tested. Seconds, and free.
  4. Note the fixture URL. Every later stage refers to it.

Stage 1 — run the review (subagent, fresh)

Spawn a subagent whose whole brief is:

Read <repo>/src/commands/review.md VERBATIM and follow it against <fixture PR url>. Wherever it says ${CLAUDE_PLUGIN_ROOT}, substitute <repo>/src — you are exercising the WORKING TREE, not the installed plugin. You have no other instructions and no knowledge of what the PR contains.

FORBIDDEN: paraphrasing the command file into the brief, hinting at the planted defects, naming the stacks involved, or telling it what a good review looks like. Every one of those invalidates the round.

Read the full file on GitHub · 113 lines

Changes

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.

  1. 8d ago First seen · 113 lines · 85 tokens per session scan A 4963ab7b0863

Subscribe to this mod's changes

e2e-loop is a skill published in the GitHub repository TOMOSIA-VIETNAM/open-pr (42 stars, last pushed 2d ago), licensed MIT. It adds 85 tokens to every session and 1,396 once invoked, about $0.0004 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.

Related

Other skills, from other repositories

playwright-stability

Make a Playwright E2E suite stable and realistic — kill flaky tests and authenticate like a real user via storageState (login once, reuse). Use when E2E tests are flaky, slow, re-login in every test, mock auth instead of using it, or when hardening a suite before relying on it.

LFTPadilla/agent-dev-kit · 70 tokens

live-qa

Exploratory QA of a running app by driving a real browser like a user via the Playwright MCP server. Use when the user wants to "QA" a feature live, walk a flow as a real user, smoke-test before a PR, or find issues that scripted specs miss. Complements deterministic Playwright specs — this is the human-like…

LFTPadilla/agent-dev-kit · 77 tokens

tdd-workflow

Use this skill when writing new features, fixing bugs, or refactoring code. Enforces test-driven development with 80%+ coverage including unit, integration, and E2E tests.

affaan-m/ECC · 43 tokens

e2e-testing

Playwright E2E testing patterns, Page Object Model, configuration, CI/CD integration, artifact management, and flaky test strategies. Use when writing Playwright tests, structuring page objects, or fixing flaky E2E runs in CI.

affaan-m/ECC · 53 tokens

atf-testing

Build ServiceNow Automated Test Framework tests and suites — impersonation, form steps, assertions, server-side script steps, test parameters, and execution via snowcreateatftest / snowexecuteatftest.

serac-labs/serac · 45 tokens

acceptance

A skill for writing, running, and repairing human-written acceptance checks stored in an `acceptance/ACCEPTANCE.md` file. Acceptance checks are executable tests of whether the finished result matches the requested outcome.

cryndoc/polisade-orchestrator · 129 tokens