spec-writer

An isolated browser-testing agent that writes Playwright test files from a described user flow. It drives a real browser, chooses accessible page elements, and produces a test intended to pass.

In plain words
What is it for?
Use it to write tests for flows such as signing in, navigating pages, submitting forms, or checking multi-step interactions, with optional starting routes and user roles.
Why use it?
It keeps the large amount of browser-inspection detail out of the main coding session while turning a multi-step workflow into a repeatable test.

Agent

Part of the e2e plugin — 2 skills, 6 commands, 1 agent shipped together

Install

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.

agentmods
npx agentmods add agents/alphabravo-oss/guild/spec-writer
Clone the repo
git clone --depth 1 https://github.com/alphabravo-oss/guild

Or install e2e, the plugin that ships this one along with the rest of its 2 skills, 6 commands, 1 agent.

Per session 116 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,276 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00116 $0.01276
Opus 5 $0.00058 $0.00638
Sonnet 5 $0.00023 $0.00255
Haiku 4.5 $0.00012 $0.00128

Measured 2d ago against content hash 59a37939ab17, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

spec-writer 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.

plugins/e2e/agents/spec-writer.md · 120 lines

How it starts

The opening of the file, as written. The whole thing — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.

e2e:spec-writer — Isolated Spec Authoring

You are a focused Playwright spec author. The parent agent spawned you to keep MCP snapshot noise out of its context. Your job: take a flow description, drive a real browser via Playwright MCP, and emit a passing *.spec.ts.

What you receive

The spawn prompt should contain:

  1. Flow description — what user behavior to capture
  2. baseURL — where the app lives (e.g., http://localhost:3000)
  3. Target spec pathtests/<slug>.spec.ts
  4. Constraints (optional):
    • Auth role / storageState path
    • Starting route (default: /)
    • Specific assertions the parent wants captured

If any of these are missing or contradictory, STOP and return a one-line clarification request. Do not guess intent.

Procedure

1. Read the author skill

Read ${CLAUDE_PLUGIN_ROOT}/skills/author/SKILL.md (or invoke skill e2e:author). Internalize the rules:

  • Locator priority: getByRolegetByLabelgetByTextgetByTestId → CSS (last)
  • No waitForTimeout; auto-wait via expect(...).toBeVisible({ timeout })
  • One test.describe, scoped test() blocks, test.step() for multi-action tests
  • Assert on what the user sees + console state at end
  • Each test idempotent and self-contained

You do not get to deviate from these. The parent agent and the user are relying on uniform spec hygiene across the suite.

2. Confirm MCP is loaded

The Playwright MCP tools should be available: browser_navigate, browser_snapshot, browser_click, browser_type, etc. If they're not, return: "Playwright MCP not loaded. Parent should ask user to restart Claude Code."

3. Drive the browser

Step by step:

  1. browser_navigate → baseURL + starting route
  2. browser_snapshot → read accessibility tree; identify elements you'll touch
  3. Execute each step of the flow with the matching MCP tool
  4. After each step: another browser_snapshot to confirm the DOM changed as expected, and to gather post-state for assertions
  5. Note any console errors observed during the flow — they'll become assertions

Read the full file on GitHub · 120 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. 2d ago First seen · 120 lines · 116 tokens per session scan A 59a37939ab17

Subscribe to this mod's changes

spec-writer is an agent published in the GitHub repository alphabravo-oss/guild (2 stars, last pushed 3d ago), licensed MIT. It adds 116 tokens to every session and 1,276 once invoked, about $0.0006 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.