shinka-convert

Convert an existing codebase in the current working directory into a ShinkaEvolve task directory by snapshotting the relevant code, adding evolve blocks, and generating `evaluate.py` plus Shinka runner/config files. Use when the user wants to optimize existing code with Shinka instead of creating a brand-new task from a natural-language description.

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

100%Weight 40%Scale 1-3

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

A well-structured, highly actionable conversion skill: it pairs concrete executable templates with a clearly sequenced workflow that includes an explicit validation/smoke-test checkpoint, and it externalizes runner/config templates into real bundled scripts with clear navigation. Minor conciseness gains are possible but it does not pad with concepts Claude already knows.

Dimension	Reasoning	Score
Conciseness	The body is lean and instructional throughout — it assumes Claude's competence, includes no padding about basic concepts, and the code/commands earn their tokens; minor redundancy in the Notes section is not enough to drop it below the level-3 'every token earns its place' anchor.	3 / 3
Actionability	It provides fully executable code templates (Python/non-Python conversion templates, evaluator template), concrete naming requirements (`initial.<ext>`, `EVOLVE-BLOCK-START`/`END`), and a copy-paste smoke-test command, matching the level-3 'fully executable, copy-paste ready' anchor.	3 / 3
Workflow Clarity	A clear 9-step sequence is given with an explicit validation checkpoint at step 8 (smoke test with confirmation criteria that the evaluator runs without exceptions and writes required outputs), so it is not capped despite touching snapshot/rewrite operations.	3 / 3
Progressive Disclosure	The body is well-organized into sections and properly signals one-level-deep bundled references (`scripts/run_evo.py`, `scripts/shinka.yaml`) that are real files; inline code templates are core instructional content rather than material better externalized, fitting the level-3 anchor.	3 / 3
	Total	12 / 12 Passed

Description

85%Weight 40%Scale 1-3

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

A strong, specific description that concretely states its actions and includes an explicit 'Use when' trigger with good disambiguation against sibling skills. Its only weakness is somewhat narrow trigger-term coverage with few natural-language variations.

Suggestions

Broaden trigger terms to capture more phrasings a user might say, e.g. 'turn my repo into a Shinka task', 'run Shinka on existing code', or 'evolve my current script'.

Consider naming a concrete artifact trigger (e.g. mentions of `evaluate.py`, `EVOLVE-BLOCK`, or `metrics.json` in the current dir) to sharpen when-this-fires.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'snapshotting the relevant code', 'adding evolve blocks', and 'generating `evaluate.py` plus Shinka runner/config files' — matching the level-3 anchor of multiple specific actions, not the vague single-domain level-2 example.	3 / 3
Completeness	It clearly states what the skill does (convert a codebase by snapshotting, adding evolve blocks, generating evaluator/runner/config) and gives an explicit 'Use when the user wants to optimize existing code with Shinka...' trigger, answering both what and when.	3 / 3
Trigger Term Quality	It includes natural phrases a user would say ('optimize existing code with Shinka', 'existing codebase'), but coverage is narrow with few variations, fitting the level-2 'some relevant keywords but missing common variations' rather than the broad multi-form list of the level-3 anchor.	2 / 3
Distinctiveness Conflict Risk	The niche is clear — converting an existing codebase — and the 'instead of creating a brand-new task from a natural-language description' clause explicitly disambiguates it from sibling skills like `shinka-setup`, giving it distinct triggers unlikely to conflict.	3 / 3
	Total	11 / 12 Passed

Validation

100%

Warnings & errors only

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.

Repository: SakanaAI/ShinkaEvolve
Path: skills/shinka-convert/SKILL.md
Commit: b67a073

Reviewed: 12 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.