Odd-e

Backend Java production code, Spring Boot controllers/services/repositories, controller return values, imports, schema ERD pointer. Use when editing backend production Java.

中文简介正在准备中。当前显示原文说明。

Backend Java tests, controller-level stable-boundary tests, JUnit, @Nested, MakeMe builders, @Transactional, parameterized tests. Use when writing or changing backend JUnit tests.

中文简介正在准备中。当前显示原文说明。

Backend Java — Spring Boot controllers/services, JUnit controller tests, MakeMe. Read this first for any backend work. Use when working on the Java Spring backend.

中文简介正在准备中。当前显示原文说明。

related to donut CLI. Use when working on donut-cli.

中文简介正在准备中。当前显示原文说明。

Set up and run the Donut dev environment on Cursor Cloud VM (no Nix). Use when working on Cloud VM, running tests on Cloud VM, or encountering command errors related to missing nix. Triggers on: cloud VM, cloud agent, no nix available, /workspace path.

中文简介正在准备中。当前显示原文说明。

Regenerate docs/database-erd.md from MySQL information_schema (Mermaid ERD: foreign keys and key columns). Use after Flyway schema changes, new migrations, or when the diagram is stale. Triggers on: database ERD, database-erd.md, export_database_erd, schema diagram, ER diagram after migration.

中文简介正在准备中。当前显示原文说明。

Database migration, Flyway, SQL migrations, schema changes, database versioning, migration files, squashing historical migrations. Use when adding or changing Flyway migrations or schema.

中文简介正在准备中。当前显示原文说明。

Use and cite this project's current Accepted Architectural Decision Records for architecture-shaped work, follow supersession, surface conflicts instead of silently drifting, and preserve human ownership of decisions and exceptions. Use when a task mentions ADRs or architectural constraints, may reverse an existing decision, or crosses architecture areas defined by this project.

中文简介正在准备中。当前显示原文说明。

Resolves a reported discrepancy, defect, or regression by gathering expected versus actual behavior and passing the report into bounded shared execution, or by placing known larger or inconclusive work first in this project's product backlog. Use for a bug report, defect, regression, broken or unexpected behavior, a mismatch between intended and actual results, or a larger or inconclusive bug that needs a first-priority story. Do not use for the word "fix" alone, a reporting-only request, or refinement-only work.

中文简介正在准备中。当前显示原文说明。

Executes one selected story or bounded retrospective correction through an executable plan, or one authorized planless slice from a selected simple story or a contextual instruction, with independent refactoring, delivery, and asynchronous CI repair. Continues settled planned work into Story Wrap Up. Use to execute a plan, run slices, execute a canonical story when the caller explicitly skips slice planning, or execute a small instruction from context without a story or plan. Also admits an accepted mission that no backlog list holds, such as a standalone review, investigation, or maintenance request, into Taken before its work starts. Does not decide story scope or quick-path eligibility. `--trunk` selects Trunk Mode; omitted mode keeps Story Branch Mode. `--one-shot` keeps selected trivial work's result for review; `--auto-land` lands it. `--skip-retro` skips the automatic planned-execution retrospective. `--replan` and `--no-replan` choose whether an oversized attempt may continue through planning.

中文简介正在准备中。当前显示原文说明。

Reviews planned, completed planless quick, or quick-to-planned execution against original intent, aggregate commits, current whole-product architecture, and tests, including after cleanup. Process review is opt-in: it runs only when the project sets `skipProcessRetrospective: false` or the request asks to include it, and needs available agent history. Use for execution retrospective, product review, or backlog recommendations from current/supplied history. Supports `--skip-process` and `--skip-product`. May plan corrections, record process findings in `DearDough.md` (500-line warning, 1,000-line ceiling, recoverable lower-priority replacement on overflow), and recommend product work. Finishes quietly when settled, surfaces unresolved attention, and never implements findings.

中文简介正在准备中。当前显示原文说明。

Dough Land lands everything in one reviewed checkout on the authorized remote trunk: commits all of its changes, reconciles and publishes them without force, and refreshes a supplied default checkout when safe. An owned worktree is retired with its branch after completion gates hold and trunk contains its work; the default checkout stays in place. Ordinary landing refuses unfinished execution before commit; `--keep-worktree` selects an explicit retained increment. Use only on explicit invocation, such as "/dough-land", "use Dough Land", "land this worktree", or "land" for changes made on the default checkout, or when another skill's validated keep instruction or findings maintenance handoff links here. A casual "keep", "looks good", or approval does not invoke it. Hosted merges and pull requests are out of scope. With `--process-retrospective`, it also reviews the landing's own process and records supported findings in `DearDough.md`.

中文简介正在准备中。当前显示原文说明。

Adopts supplied, approved DD-NNN to ODF-NNN finding-name mappings in this project's DearDough.md while preserving evidence and former-code traceability. Use only on explicit skill invocation or an explicit request to adopt or rename finding codes; encountering findings or mapping recommendations does not authorize maintenance. Other finding cleanup and status changes are out of scope.

中文简介正在准备中。当前显示原文说明。

Plans bounded manual or exploratory observation of requested product behavior when the developer explicitly requests manual testing, exploratory testing, or an active plan slice requires it. Covers web, CLI, API, desktop, or combined surfaces. Does not run proactively or replace required automated tests.

中文简介正在准备中。当前显示原文说明。

Finds and assesses existing solutions across the whole product before an implementation responsibility is added or relocated. Use for PFE, Proudly Found Elsewhere, or when deciding whether to reuse, change, modularize, or replace an existing product solution.

中文简介正在准备中。当前显示原文说明。

Refactors concepts implicated by the current uncommitted change before commit. Includes untouched representations and orchestration needed for coherence, stops before unapproved cross-subsystem refactoring or disputed plan behavior, and tests only after edits. Use after a slice, for post-change refactoring, or to clean up the current change.

中文简介正在准备中。当前显示原文说明。

Maintains and reprioritizes a product backlog list of canonical story or bounded-correction references, and records necessary blocking story dependencies in canonical homes. Use to add, take, reorder, or complete backlog items, adopt or repair the done catalog beside them, report which story identities are already used before allocating a new one, record or inspect a justified prerequisite, correct an evidenced supplier agreement after cleanup, or resolve product backlog merge conflicts. Excludes classroom and workshop exercise backlogs.

中文简介正在准备中。当前显示原文说明。

Resplits a large story after slice-plan refinement into smaller stories and mapped slice plans. Use when resplitting is requested, including after a more-than-15-slices recommendation. Leaves the first story and plan refined; later plans await story refinement and backlog priorities are reconsidered.

中文简介正在准备中。当前显示原文说明。

Refines an executable plan in place into proportionate, proof-owned Behavior/Structure slices. Use during or after dough-slice-planning when boundaries fragment cohesive work, combine independent concerns, accumulate special cases, or overrun. Creates no new plan and preserves the selected story or bounded correction outcome. After rewriting the plan, reassess readiness through the shared preparation procedure without granting Take or execution.

中文简介正在准备中。当前显示原文说明。

Plans one understood, bounded story, retrospective correction, or remaining work from an understood instruction as an executable sequence of Behavior/Structure slices with outside-in proof and safe stopping points. Use when a selected story is ready for implementation planning, an execution retrospective has one bounded correction to plan, or authorized remaining work from a sufficient instruction needs an ordinary plan. Stays within the triggering instruction's execution authority: finish after writing the plan and recording its readiness unless that instruction explicitly also requests execution. Invokes slice-plan refinement when remaining plan concerns can be resolved within the understood outcome, then records remaining concerns in the plan and readiness through the shared preparation procedure without granting Take or execution.

中文简介正在准备中。当前显示原文说明。

Challenges and decomposes a broad product problem into ordered Valuable/Visible/Vertical stories with rough effort hypotheses. Use when the parent problem, candidate stories, or learning priority are unresolved. Writes one non-executable seed.

中文简介正在准备中。当前显示原文说明。

Clarifies selected stories before execution planning by establishing goal, scope, and key examples in each story's seed, distinguishing promises from rejection constraints. Adds UI or architectural detail only when needed. Continues a ready story into slice planning in the same workspace; `--refine-only` stops at the refined story. `--one-shot` refines a queued story without publishing an assignment and keeps the result for review, or lands it with `--auto-land`. Use for selected-story refinement, not broad problem decomposition, candidate selection, or slice sizing.

中文简介正在准备中。当前显示原文说明。

Closes one completed feature story, bounded retrospective correction or context-only planless execution using available execution context. Assimilates lasting product knowledge, settles existing follow-ups and authorized product decisions, and removes spent plan and history with Git recovery. Reports missing or unfinished inputs truthfully. Use to close completed work, delete spent records, or continue settled planned execution under Execute Plan's continuation rule. With `--process-retrospective`, it also reviews the wrap-up's own process and records supported findings in `DearDough.md`.

中文简介正在准备中。当前显示原文说明。

Speeds up local test feedback through profiling and measured experiments across related tests. Use to optimize slow tests or test-suite performance. Preserves behavioral coverage and confidence while pursuing fewer redundant cases, less code, and cleaner design. With --resolve, triages recorded candidates without profiling or optimizing.

中文简介正在准备中。当前显示原文说明。