medgemma-setup

Interactive setup wizard for Med-Guide. Guides non-technical users through the complete setup from scratch — editor configuration, uv, Modal, HuggingFace, deployment, and first analysis. Use when the user first opens the project, says "setup", "install", "kurulum", "get started", "start setup", or when .env file is missing.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Quality

Content

77%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.

The content is highly actionable with a clear, well-validated multi-step workflow and strong error-recovery feedback loops. Its main weaknesses are verbosity from inlined bilingual message templates and a monolithic structure that underuses progressive disclosure.

Suggestions

Move the verbatim English/Turkish message templates (STEP 0, STEP 10, HELP) into a separate references file and reference it by language, eliminating duplicated inline blocks to improve conciseness.

Extract the per-error recovery branches (STEP 6/STEP 8 failure cases) into a references/error-recovery.md so the main flow stays lean while preserving the feedback loops.

Templating the "When the user says 'done' → go to STEP N" transitions into a compact state table would cut repeated phrasing without losing workflow clarity.

Dimension	Reasoning	Score
Conciseness	The body is mostly actionable instruction rather than concept explanation, but full English+Turkish verbatim message blocks (STEP 0, STEP 10, HELP) and repeated "When the user says done" phrasing add noticeable padding that could be templated.	3 / 5
Actionability	Provides fully executable, copy-paste-ready commands (uv sync, curl install, modal deploy scripts/modal_medgemma.py), exact URLs, exact .env contents, and specific per-error fixes covering the common cases.	5 / 5
Workflow Clarity	STEPS 0–10 are explicitly sequenced with validation checkpoints ("Done?", re-check commands like modal profile current / modal secret list) and feedback loops for recovery (deploy fails → return to STEP 5; uv missing → restart editor → retry).	5 / 5
Progressive Disclosure	Step-based section structure is clear and project-file references are signaled, but no skill bundle files exist and the 619-line file inlines large bilingual message templates and error-recovery content that would be better split into separate reference files.	3 / 5
	Total	16 / 20 Passed

Description

95%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.

The description is strong: it clearly states what the skill does, when to use it, and names the concrete setup stages with a comprehensive set of natural trigger phrases including a multilingual synonym. Minor tightening of the action framing could lift specificity from good to fully comprehensive.

Dimension	Reasoning	Score
Specificity	Enumerates concrete setup stages — "editor configuration, uv, Modal, HuggingFace, deployment, and first analysis" — but folds them under a single guiding action rather than listing multiple distinct verbs, leaving a minor coverage gap versus the comprehensive anchor.	4 / 5
Completeness	Explicitly answers both what ("Interactive setup wizard for Med-Guide... complete setup from scratch") and when ("Use when the user first opens the project, says...") with concrete trigger phrases.	5 / 5
Trigger Term Quality	Natural trigger phrases are comprehensive: "setup", "install", "kurulum", "get started", "start setup", plus the state trigger ".env file is missing", including a non-English synonym and a file-based condition.	5 / 5
Distinctiveness Conflict Risk	Targets a clear niche — Med-Guide project setup with a specific tool stack (uv, Modal, HuggingFace, MedGemma) — with distinct triggers and minimal overlap risk with unrelated skills.	5 / 5
	Total	19 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (618 lines); consider splitting into references/ and linking	Warning
referenced_paths_exist	Referenced path issues: 5 missing	Warning

	Total	14 / 16 Passed

Repository: burakcanpolat/med-guide
Path: .agents/skills/medgemma-setup/SKILL.md
Commit: 83a6c1e

Reviewed: 1 day 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.