Skip to content

ilang-ai/autocode / skills/product-study

product-study

Unverified

Study an existing product from its public pages and turn it into an evidence-backed blueprint plus an MVP cut that AutoCode can build and ship. Use this whenever the user wants something "like X", "similar to X", "my own version of X", "a clone of X", names a competitor they admire, or pastes a product URL and asks to build it — even when they never say "research" or "analyze". Public sources only (official site, docs, pricing, changelog, public repos, app-store listings, public reviews); no logins, captures, decompiling, or copying of code, copy, images or brand.

references/blueprint-template.md

View on GitHub
# Product blueprint: TODO product name

<!--
Write the body in the user's language. Keep the numbered headings exactly as
they are (the number is what `blueprint.py check` keys on; the words after it
may be translated). Delete every HTML comment and every TODO before `check`.

Claim lines: in sections 1-8 every bullet and every table row is a claim and
must end with a source tag and a confidence tag, for example
  - Issues can be grouped into cycles of fixed length [S12] {high}
  | Workspace -> Team -> Project -> Issue | S12, S13 | high |
Confidence: {high} = stated verbatim by an official page; {medium} = official
page implies it, or a credible third party states it; {guess} = your inference,
no page says it. {guess} lines may omit the source tag; the others may not.
Chinese tags {高} {中} {推测} are accepted too.
-->

Product: TODO name · Official site: TODO url · Studied on: TODO date · Goal: TODO one sentence on what the user wants to build and for whom

Reading guide: this blueprint describes **structure** (what the product does, how its objects relate, what its users go through), gathered only from public pages listed under Sources. It is a map for building your own product. It does not carry the original's interface, copy, images or brand, and nothing in it is copied from a non-public source.

## 0. Summary

<!-- one paragraph: what it is, who it is for, the one thing it does best, the pricing shape. Plain words, no claims that are not in a source. -->

TODO

## 1. Feature inventory

<!-- Every feature you can establish from the sources. Tier: core = the product is unusable without it; plus = valuable but separable; drop = you would not build it in your version. Group by area if there are more than ~15. -->

| Feature | What it does (one line) | Tier | Sources | Confidence |
|---|---|---|---|---|
| TODO | TODO | core | S1 | high |

## 2. Information architecture

<!-- The nesting of containers and navigation as the user sees it (Workspace > Team > Project > Issue; sidebar sections; top-level pages). Bullets or a short tree in a code block. -->

- TODO [S1] {high}

## 3. Core objects and data model (inferred)

<!-- Entities, their key fields, and relations. Most of this is inference from docs and screenshots, so expect {guess} and {medium}. Mark 1:N / N:M. -->

| Object | Key fields | Relations | Sources | Confidence |
|---|---|---|---|---|
| TODO | TODO | TODO | S1 | medium |

## 4. Key user flows

<!-- 3 to 6 flows that define the product (onboarding, the main create/edit loop, collaboration, reporting). Numbered steps; each flow is one bullet with sub-steps or a short list. -->

- Flow A: TODO -> TODO -> TODO [S1] {medium}

## 5. Technical stack clues

<!-- Only what public pages say: API style (REST/GraphQL), SDKs, webhooks, auth model, platforms, job postings naming languages/frameworks, open-source repo languages. No guessing from bundle names or network traffic: that is out of scope. -->

- TODO [S1] {high}

## 6. Pricing and business model

<!-- Plans, what gates each plan, free tier limits, per-seat vs usage, enterprise extras. -->

| Plan | Price | What it unlocks | Sources | Confidence |
|---|---|---|---|---|
| TODO | TODO | TODO | S1 | high |

## 7. Integrations and ecosystem

<!-- Native integrations, public API, import/export, marketplace, CLI, mobile/desktop apps. -->

- TODO [S1] {high}

## 8. Gaps and differentiation opportunities

<!-- Where public reviews, comparison pages, roadmap or the docs themselves show a hole, a complaint, a market the product does not serve. This is where your product gets its reason to exist. Inference is fine, tag it {guess}. -->

- TODO {guess}

## 9. MVP cut

<!--
The hand-off to the build phase. Only the core tier survives, sized for what an AI agent can build and deploy in one session:
  at most 5 core objects, at most 7 screens, no third-party integrations in v1, one auth method, one deployment target.
Each line is a decision, not a claim: no source tags needed. Keep the user's goal from the header in view; the MVP serves their users, not the original's.
Keep the bold labels starting with the English words Objects / Screens / Acceptance / Learned (add a translation after them if you write in another language):
the gate finds them by those words. List items separated by commas or 、; a comma inside parentheses does not split an item.
-->

**Users and the one job:** TODO who uses v1 and the single job it must do

**Objects (<= 5):** TODO

**Screens (<= 7):** TODO

**Dropped from v1 and why:** TODO

**Build order (each step ends in something the user can click):**
1. TODO
2. TODO
3. TODO

**Acceptance (v1 is done when):** three to five lines, each in three parts separated by `->`: what the user does -> what they should see -> what a machine can check (write `manual` if nothing can)
1. TODO user does -> should see -> machine checks
2. TODO user does -> should see -> machine checks
3. TODO user does -> should see -> machine checks

**Learned vs own (学来的 / 自己定的):** TODO the rules and numbers taken from the target, each with its [S#]; one line saying everything else (look, layout, copy, defaults) is our own design

## 10. What we will not copy

- Interface look, layout, colors, icons and typography of the original: we design our own.
- Copy, screenshots, images, logo, product and feature names that are trademarks.
- Any code from the original's repositories or bundles. (If a studied repository is copyleft, say so here: its architecture informed this blueprint; none of its code is used.)
- Anything that was not on a public page: no account areas, captures, decompilation or leaked material were used.

## Sources

<!-- Generated by `blueprint.py sources <slug>` from sources.json. Do not edit by hand. -->
rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).Stars: 198kUpdated Claude CodeCodexUnverifiedInstall
doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.Stars: 198kUpdated Claude CodeCodexVerifiedInstall