# Drydock UAT Kits A **kit** is one known project that Drydock rebuilds unattended, end to end, against a real model: `init` → `import` → `analyze` → `plan` → `build`, then scored and sealed into a self-verifying proof kit. Kits are Drydock's full-suite acceptance capability and its public worked examples. Each kit directory is self-contained and self-runnable. `uat/Toml/` is published on its own as `drydock-example-toml`; nothing outside it is needed to reproduce a run. ## Layout ```text uat/ README.md this file / one kit, publishable as its own repository README.md what the kit builds and how to run it uat.json source bundle, updates, and test command index.html landing page linking every run inputs/ optional lifecycle decisions seeded before analysis sources/ Blueprint inputs and supplied build assets updates/ replacement sources that drive incremental rebuilds runs// one complete unattended run ``` `runs/` is generated. It is ignored by Git in this repository and published deliberately in an example repository. ## Running ```bash drydock uat # every kit under uat/ drydock uat Toml # one kit drydock uat --report # rebuild proof kits from completed runs drydock uat --report Toml # rebuild one kit's proof kits ``` | Flag | Effect | |---|---| | `--uat-root ` | Directory holding the kits (default `/uat`) | | `--max-build-passes ` | Repair passes allowed per build before the kit fails | | `--llm-provider`, `--model`, `--effort` | Provider and model selection for the whole run | A run is long and consumes subscription quota. The `Toml` kit takes roughly thirty minutes and eighteen LLM calls. ## Kit definition `uat.json` declares: - `target` — the Target name created by `init`. - `sources` — kit-local files imported before the initial lifecycle. Required and nonempty. Paths are relative to the kit root; the bundle is flattened to `sources/` in the build, so basenames must be unique. - `updates` — files that replace an imported basename to drive `import --update` → `refit --sources` → incremental build, once each, in order. - `sea_trials` — optional kit-local Sea Trials path, seeded after `init` and before `analyze`. - `technology_stack` — optional kit-local technology-stack path, seeded at the same point. Named Rigging files are validated against the catalog during discovery. - `test_command` — argv run from the completed application root after the build. A nonzero exit fails the kit. Both lifecycle inputs are explicit. Root-level magic filenames are ignored. When either key is omitted, `analyze` creates that artifact inside the run Target. Every non-Markdown source is an artifact for the build to use — a test harness, a fixture corpus, a tool — and is staged verbatim into the build directory's `sources/`. Markdown is specification prose and becomes Blueprint input instead. ## Reading a run ```text runs// README.md run verdict, elapsed time, token and cost accounting index.html linked proof kit result.json every child command, argv, exit code, elapsed time SHA256SUMS integrity manifest inputs/ exact declared lifecycle inputs for this run sources/ the exact bundle imported for this run workspace/ the isolated Drydock workspace, including targets// build// the delivered application evidence/ commands/ stdout and stderr of every child command prompts/ the assembled prompt for every LLM call prompt_outputs/ the parsed model output provider_raw/ the unmodified provider transcript llm.jsonl one record per call: tokens, elapsed time, execution id manifest.json evidence index ``` When a build fails, the authoritative diagnosis is `workspace/targets//evidence/.md`: the pre-build acceptance observation, the stacked context, the build-directory changes, and the post-build result for every criterion. Verify a kit with: ```bash cd runs/ && sha256sum -c SHA256SUMS ``` Absolute paths are rewritten to run-relative paths when the kit is sealed, so a run stays readable on any machine.