Amadeus feedback optimization

Skill amadeus-dlc/amadeus/dist/kimi/.kimi-code/skills/amadeus-feedback-optimization

Agent skills, templates, examples, validators, and docs for operating Amadeus DLC, a lifecycle contract for AI driven software development.

Install

npx -y skills add amadeus-dlc/amadeus --skill amadeus-feedback-optimization

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

5 stars5 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Run the AI-DLC `feedback-optimization` stage (operation phase) in isolation, without advancing the main workflow. Packages `/amadeus --stage feedback-optimization --single`: the engine emits one run-stage directive for feedback-optimization and its gate, the conductor runs it, then the single-stage run commits a synthetic-id pair and stops. The main workflow's Current Stage is never touched.

SKILL.md

1.7 KB, 294 tokens by cl100k_base, as published. Nobody here has run it

AI-DLC Stage Runner — feedback-optimization

Run the feedback-optimization stage on its own. This is opt-in packaging over /amadeus --stage feedback-optimization --single; the same stage is always reachable via that flag without this skill.

Steps

Ask the engine for the single-stage directive:
```
bun .kimi-code/tools/amadeus-orchestrate.ts next --stage feedback-optimization --single
```
The engine emits one run-stage directive for feedback-optimization (carrying the lead agent, the resolved consumes/produces paths, the rules and sensors in context, and — on this first directive — the conductor persona). Run the stage exactly as the directive describes; do not load the conductor persona by hand, the engine delivers it.
When the stage's work is done, commit the single-stage record:
```
bun .kimi-code/tools/amadeus-orchestrate.ts report --single --stage feedback-optimization --result completed
```
This records a STAGE_STARTED / STAGE_COMPLETED pair under a synthetic workflow id and stops. It NEVER writes the main workflow's Current Stage — a single-stage run is isolated by design (the tool refuses to advance the main workflow).

Gives 0 of the 12 instructions most performance cost skills give in 294 tokens

Counted across 803 of the 1,058 authors here whose files we hold, read 2026-08-07

Keep skill files under 500 lines or tokensin 82 of 803, across 16 files
Use imperative form in instructionsin 80 of 803, across 9 files
Draft assertions while test runs are in progressin 75 of 803, across 9 files
Create two to three realistic test promptsin 74 of 803, across 9 files
Write skill descriptions to be pushyin 72 of 803, across 7 files
Save test cases to evals JSONin 72 of 803, across 6 files
Ask questions about edge cases and input formatsin 72 of 803, across 7 files
Save timing data immediately when runs completein 70 of 803, across 5 files
Include all trigger conditions in the skill descriptionin 69 of 803, across 3 files
Launch all test runs in a single turn or simultaneouslyin 69 of 803, across 3 files
Capture intent before writing a skillin 67 of 803, across 1 file
Import directly instead of barrel filesin 52 of 803, across 15 files

Said here and by no other author read

run the stage exactly as the directive describes
commit the single-stage record when stage work is done

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.