CtrlK
BlogDocsLog inGet started
Tessl Logo
PluginPublic

ai-unified-process/aiup-flyway

AI Unified Process migration layer - Flyway schema migrations from the entity model

77

Quality

97%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Overview
Quality
Evals
Security
Files

Quality

Content

92%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-crafted procedural skill body: executable SQL examples, exact conventions, a validated workflow with a fix-and-retry loop, and clean one-level-deep references that all resolve to real files. The single weakness is verbosity in its rationale passages, which slightly inflates token cost without adding executable guidance.

Suggestions

Trim the multi-line rationale passages (e.g. 'A migration that never runs is worse than none...' and the column-by-column spread argument) to one-sentence justifications — the rules they support are already stated.

The Instructions opening largely restates the frontmatter description; compress it to the input contract and the entity-not-column rule to cut duplicate tokens.

DimensionReasoningScore

Conciseness

The body is dense with project-specific, non-obvious rules (Spring Boot 4's flyway starter split, Flyway 10 database modules, checksum-based immutability of applied scripts) and never explains basics, so it is well above the midpoint. It is not 5 because several multi-line rationale passages ("A migration that never runs is worse than none...", the column-by-column spread argument, which partially repeats the frontmatter) could be tightened to one line each without losing guidance.

4 / 5

Actionability

Fully executable guidance throughout: a copy-paste-ready example migration (CREATE SEQUENCE / CREATE TABLE with CHECK, UNIQUE, and a marker comment), exact naming patterns ("{table_name}_seq", "V001__create_room_type_table.sql"), a concrete SQL marker snippet, and a fill-in Handover template. It is not 4 because the concrete examples cover the common cases end to end.

5 / 5

Workflow Clarity

The 9-step workflow has an explicit validation checklist in step 7, an executed build check, and a genuine feedback loop — "When a check or the build fails, fix the new migration — it has not run anywhere yet — and repeat this step until both pass" — plus destructive-operation guards (drops require user confirmation). Database operations get the validation loop the rubric requires, so it is not 4.

5 / 5

Progressive Disclosure

Both referenced files (references/layer.md, references/markers.md) exist, are exactly one level deep, and are clearly signaled with what they contain and when to consult them; the Markers-table rows are correctly split out of the body, and the two external documentation links are well placed. The body retains only execution-time rules, so nothing that belongs in a bundle file is inlined — matching the 5 anchor rather than 4.

5 / 5

Total

19

/

20

Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

An exceptionally strong description: concrete capabilities, explicit third-person voice, comprehensive natural trigger phrases with synonyms, and explicit boundary marking against test and persistence skills. Its only weakness is density — the policy detail about complete tables and ALTER TABLE duplicates body content and could be trimmed for scannability without losing trigger value.

DimensionReasoningScore

Specificity

The description names multiple concrete, artifact-level actions — "versioned V*.sql scripts with sequences, tables, snake_case columns, CHECK and UNIQUE constraints, and foreign keys", "ALTER TABLE only where an existing table lags behind the entity model", and "Marks a constraint that enforces a business rule with -- UC-XXX BR-YYY" — giving comprehensive coverage of what the skill produces. It is not 4 because the action list is exhaustive rather than having minor gaps.

5 / 5

Completeness

The "what" is explicit and detailed (builds the migration layer: complete tables for new entities, ALTER TABLE for lagging ones, business-rule markers) and the "when" is an explicit 'Use when' clause with concrete quoted trigger phrases and scope limits ("in a project that combines layer plugins"). Both what and when are answered explicitly with triggers, matching the 5 anchor.

5 / 5

Trigger Term Quality

Trigger coverage includes natural user phrasings ("create a migration", "write a Flyway migration", "add a column", "set up the database schema") plus synonyms ("schema migration, DB migration, database versioning, or SQL migration files") and the V*.sql file pattern. It is not 4 because both natural phrases and synonym variations are comprehensively present.

5 / 5

Distinctiveness Conflict Risk

It carves a clear niche (Flyway schema-migration layer of a layered plugin process) and actively disambiguates against neighbors — "Test data migrations belong to the test skills" and "Never touches: the persistence, service, API, and UI layers". It is not 4 because the exclusions leave minimal overlap risk with adjacent skills.

5 / 5

Total

20

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Reviewed

On this page