AI Flow

@@ -1,5 +1,17 @@ # AI Flow +## AI Flow is a dynamic skill++A **dynamic skill** is a small Rust binary that holds the AI's hand through a long job. It says which step comes next, hands over ready-made code for the routine parts, checks the AI's work with deterministic linters, and answers every command with hints. AI Flow and every `adom-aiflow-<stage>` app (ordering, probing, video, ...) are dynamic skills, and they all follow the same rules:++- **The AI does about 97 % of the work. The binary is about 3 %.** The binary is the guide and the checker. The thinking, the design decisions and the actual work belong to the AI.+- **A guide, never a gate.** No step is a hard gate. A dynamic skill suggests what to do next; it does not decide whether the AI may continue.+- **Linters give feedback, gently.** A check reports what it found and suggests a next action. It is a suggestion, not a verdict: the AI and the human decide what to do with it.+- **It never tries to do the AI's job.** AI keeps getting better. A binary that does the AI's thinking for it, or walls it in with rules, is out of date the day a better model ships. Ideas and feedback stay useful; hard rules do not.+- **The only error is "I could not run"** (a missing file, a tool that is not installed), and even that answer says what to try next.++If you build or change a flow app, hold it to this. A new requirement, a mandatory field or a step that refuses is a sign the binary has started doing the AI's work.+ **Open source under the [MIT License](LICENSE).** The complete AI Flow source is available in this wiki repository: [Rust workspace](Cargo.toml), [CLI source](crates/adom-aiflow/src/main.rs), and [package metadata](package.json). Third-party dependencies and assets retain their own licenses and notices.  ```bash

@@ -8,6 +8,18 @@ description: >-  This tool is the place-route-pour-analyze skill made executable: the step order is a file, the checks report what is wrong, the hints arrive as the answer to the command you just ran, and the ledger records what you did. Follow the commands and the skill follows itself. +## What a dynamic skill is (read this first)++adom-aiflow, and every `adom-aiflow-<stage>` app beside it, is a **dynamic skill**: a small Rust binary that holds your hand through the job. It tells you which step comes next, gives you ready-made code for the routine parts, checks your work with deterministic linters and answers every command with hints.++- **You are about 97 % of the flow. The binary is about 3 %.** It is the guide and the checker; the thinking and the work are yours.+- **It is never a hard gate, at any step.** What it prints is an idea for what to do next. You may take a different route when you have a better one; say what you did and why.+- **Linter feedback is a gentle suggestion.** A finding tells you what the check saw and offers a next action. Weigh it, fix what is real, and carry on. You and the human decide.+- **It does not do your job.** AI keeps getting better, so the binary stays out of the way of your judgement instead of encoding today's limits as rules.+- **Its only error is "I could not run"**, with what to try next.++When you work on AI Flow itself or on a flow app, keep it this way: add hints and checks that report, never requirements, mandatory fields or steps that refuse.+ **AI Flow is a suggestion engine. It never refuses.** Every check reports what it found (DRC errors stay prominent) and suggests a next action; every command completes; you and the human decide. `step` suggests the order and notes a step you skip; `finish` is a report card (what passed, what is still open, what it suggests), not a gate.  **AI Flow is a helper, not a wall.** It gives you ready-made tools (the router, pours, DRC and analysis gates, linters) so you don't write them from scratch, hints for the next step, and the Adom habits: public wiki pages for new parts that every Adom user then shares, and a project page with the 3D, schematic and board viewers and the datasheets. You are still the engineer. When a check refuses a board that is right, because AI Flow has not seen a design like it, don't stop and don't fake a pass: improve AI Flow (see "Contribute to AI Flow" at the end) and finish the board with your build.

@@ -8,6 +8,18 @@ description: >-  This tool is the place-route-pour-analyze skill made executable: the step order is a file, the checks report what is wrong, the hints arrive as the answer to the command you just ran, and the ledger records what you did. Follow the commands and the skill follows itself. +## What a dynamic skill is (read this first)++adom-aiflow, and every `adom-aiflow-<stage>` app beside it, is a **dynamic skill**: a small Rust binary that holds your hand through the job. It tells you which step comes next, gives you ready-made code for the routine parts, checks your work with deterministic linters and answers every command with hints.++- **You are about 97 % of the flow. The binary is about 3 %.** It is the guide and the checker; the thinking and the work are yours.+- **It is never a hard gate, at any step.** What it prints is an idea for what to do next. You may take a different route when you have a better one; say what you did and why.+- **Linter feedback is a gentle suggestion.** A finding tells you what the check saw and offers a next action. Weigh it, fix what is real, and carry on. You and the human decide.+- **It does not do your job.** AI keeps getting better, so the binary stays out of the way of your judgement instead of encoding today's limits as rules.+- **Its only error is "I could not run"**, with what to try next.++When you work on AI Flow itself or on a flow app, keep it this way: add hints and checks that report, never requirements, mandatory fields or steps that refuse.+ **AI Flow is a suggestion engine. It never refuses.** Every check reports what it found (DRC errors stay prominent) and suggests a next action; every command completes; you and the human decide. `step` suggests the order and notes a step you skip; `finish` is a report card (what passed, what is still open, what it suggests), not a gate.  **AI Flow is a helper, not a wall.** It gives you ready-made tools (the router, pours, DRC and analysis gates, linters) so you don't write them from scratch, hints for the next step, and the Adom habits: public wiki pages for new parts that every Adom user then shares, and a project page with the 3D, schematic and board viewers and the datasheets. You are still the engineer. When a check refuses a board that is right, because AI Flow has not seen a design like it, don't stop and don't fake a pass: improve AI Flow (see "Contribute to AI Flow" at the end) and finish the board with your build.

@@ -5,6 +5,18 @@ to a finished, routed, DRC-clean board. Every later stage (ordering from a fab, own flow app built from this template. Each one is a separate wiki page and a separate Linux CLI, owned by whoever builds it. +## What a flow app is: a dynamic skill++Every flow app is a **dynamic skill**: a small Rust binary that holds the AI's hand through one stage. It says which step comes next, ships the routine parts as ready-made code, checks the AI's work with deterministic linters and answers every command with hints.++- **The AI does about 97 % of the work. Your binary is about 3 %:** the guide and the checker, never the worker.+- **Never a hard gate, at any step.** `next` suggests; it does not decide whether the AI may continue.+- **Linters report and suggest, gently.** A finding plus a suggested next action. The AI and the human decide.+- **Do not do the AI's job.** AI keeps getting better; a binary that encodes today's limits as rules is out of date with the next model.+- **The only error is "I could not run"**, with what to try next.++Where the rules below ask the AI to check with the human first (money, publishing), that is the AI asking a person, not the binary blocking the AI.+ ## Rules  - **Its own app.** One wiki page, one Cargo project, one binary named `adom-aiflow-<stage>`. Copy this folder and