plan

A method for breaking a broad task into ordered steps with checkpoints. Each step describes a user-facing objective and a condition for considering it complete.

In plain words
What is it for?
Planning workflows such as logging in, searching, checking out, testing several features, or handling tasks whose data and state carry from one step to the next.
Why use it?
It makes multi-step work easier to follow and verify, especially when later actions depend on earlier results. It also avoids unnecessary planning for a single simple action.

Skill for Claude CodeCodex

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 skills/migoxlab/webqa-agent/plan
Any agent
npx skills add MigoXLab/webqa-agent --skill plan
Clone the repo
git clone --depth 1 https://github.com/MigoXLab/webqa-agent

Made for: Claude Code, Codex.

Per session 12 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 922 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 $0.00012 $0.00922
Opus 5 $0.00006 $0.00461
Sonnet 5 $0.00002 $0.00184
Haiku 4.5 $0.00001 $0.00092

Measured 2d ago against content hash 6a4d2fb5ab1b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

plan 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.

webqa_agent/executor/flash/skills/plan/SKILL.md · 121 lines

How it starts

The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Plan Skill

Structure your approach before acting. Decompose broad objectives into sequenced steps with completion criteria at key milestones.

When to Use

  • The task describes a multi-step workflow (login, search, checkout, etc.)
  • The objective is broad ("test fundamental functionalities")
  • You are unfamiliar with the page structure
  • The task involves state that carries across steps (form data, cart items)

Skip this skill for single-action tasks ("click the login button").

Planning Phases

Phase 1: Understand the Objective

  • What is the end goal? What does success look like?
  • Which pages/features are involved? Ignore unrelated ones.
  • What data flows through the workflow? (e.g., search query -> results -> detail page)

Phase 2: Sequence the Steps

Order steps as a continuous workflow, each following from the previous outcome.

  • Aim for 4-8 steps for substantial tasks, fewer for simple ones. Each step is one reasoning unit — typically 1-3 tool calls that achieve a single sub-goal.
  • Use semantic intent, not element selectors:
    • Good: "Click the 'Submit' button below the form"
    • Bad: "Click element #btn_47" or "Click at coordinates (400, 300)"
  • For each step, state what and why:
    • Good: "Enter a search query to test the search flow"
    • Bad: "Type 'hello world' in the search box"
  • Mark dependencies between steps: if step 4 depends on data from step 2's outcome, say so explicitly.
    • Example: "Step 4 uses the product name found in step 2's results"

Phase 3: Place Completion Checkpoints

Insert a checkpoint every 3-5 steps at key milestones.

For each checkpoint, define what to verify (the completion condition), not how to verify it:

  • Good: "Verify: search results page loaded with matching items"
  • Bad: "take_snapshot and check for a results container"

The completion condition answers: what state proves this milestone was reached? The execution cycle (system prompt Step 4) handles the how.

Read the full file on GitHub · 121 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 · 121 lines · 12 tokens per session scan A 6a4d2fb5ab1b

Subscribe to this mod's changes

plan is a skill published in the GitHub repository MigoXLab/webqa-agent (228 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 12 tokens to every session and 922 once invoked, about $0.0001 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

gtplanner

使用 GTPlanner 生成项目规划和技术架构文档。当用户需要: (1) 生成 PRD 或产品需求文档 (2) 规划项目架构 (3) 技术选型和推荐 (4) 创建详细设计文档 (5) 分析项目可行性 时触发此 skill。支持中英日西法五种语言。 触发词:"PRD"、"项目规划"、"架构设计"、"技术方案"、"设计文档"。.

OpenSQZ/GTPlanner · 114 tokens

document-query

Use when reading, extracting, summarizing, comparing, or answering questions over local or remote documents, code files, PDFs, Office files, HTML/text files, large text-heavy files, and fallback OCR for document images or scans when vision tools are unavailable or insufficient.

agent0ai/agent-zero · 56 tokens

browser-qa

Drive a real headless browser against a locally-served web UI and capture screenshot evidence, reporting step-level PASS/FAIL/BLOCKED with extracted DOM text and console errors. Use when a phase on the docker QA image must exercise a rendered UI and attach screenshots — it complements the text-evidence verify/qa-test…

nearform/lastlight · 69 tokens

qa-test

Run an automated QA flow against a CLI or a locally-served app and report step-level pass/fail with evidence. Use when asked to QA-test a feature, exercise a flow end-to-end, or smoke-test what a PR changed.

nearform/lastlight · 51 tokens

use-mobile-device

Drive iOS and Android devices using the mobile-device-mcp tools (screenshot, tap, uitree, scroll, typetext, runcode, etc.). Use when MCP device-control tools are available and the user wants to interact with a mobile device.

srmorete/mobile-device-mcp · 55 tokens

ship-pr

Ship one pull request, monitor hosted CI and review feedback, follow PR policy, and request authorization before merging and cleaning up.

conn-castle/agent-layer · 28 tokens