oma-image

Multi-vendor AI image generation with authentication-aware parallel dispatch. Routes to Codex (gpt-image-2 via ChatGPT OAuth) and Pollinations (flux/zimage, free with signup). Gemini provider is present but disabled by default (requires billing). Use for image generation, image creation, visual asset generation, and AI art.

Quality

78%

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

Fix and improve this skill with Tessl

tessl review fix ./benchmarks/runs/oma/.agents/skills/oma-image/SKILL.md

Quality

Content

81%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 highly actionable, well-sequenced workflow with excellent validation and recovery guidance; the commands are copy-paste ready and the clarification protocol is exemplary. The main drag is meta-template scaffolding that pads token cost without aiding execution, plus a small amount of content that belongs in the referenced resource files.

Suggestions

Cut the abstract scaffolding sections — 'Intent signature', 'Control-flow features', the 'Actions | SSL primitive' table, and the 'Resource scope' table — since they restate information already covered by 'When to use' and the Guardrails, and assume Claude cannot infer what it can.

Deduplicate the resource listing: the 'Dependencies', 'References', and trailing bullet list each repeat the same four resources/*.md files; keep one clearly-labeled section.

Move the host-specific reference-image path details (Claude Code image-cache, Antigravity upload dirs) and the long vendor-strategy table into resources/vendor-matrix.md and resources/prompt-tips.md, keeping only the dispatch rules inline.

Dimension	Reasoning	Score
Conciseness	The Invocation, Guardrails, and Clarification Protocol sections are dense and useful, but abstract scaffolding — "Intent signature", "Control-flow features", the `SSL primitive` Actions table, the "Resource scope" table, and a "Scheduling/Goal" header that duplicates "When to use" — is explanation Claude does not need. This is more than the 'minor instances' of over-explanation at 4, but the bulk of the content is efficient.	3 / 5
Actionability	Fully executable throughout: complete CLI syntax (`oma image generate "<prompt>" --vendor auto --size auto --quality auto --format json`), copy-paste reference-image examples (`-r ~/Downloads/otter.jpeg "same otter in dramatic lighting"`), a concrete exit-code table, named env vars, and doctor/list-vendors commands.	5 / 5
Workflow Clarity	The PREPARE→ACQUIRE→ACT→VERIFY→FINALIZE sequence is explicit, with an inline Clarification Protocol checklist, a per-failure recovery section (missing auth, outdated CLI → `oma update`, timeouts, exit codes), defined exit conditions, and a pre-submit checklist reference. Validation and feedback loops are explicit.	5 / 5
Progressive Disclosure	References are clearly signaled and one level deep (resources/execution-protocol.md, vendor-matrix.md, prompt-tips.md, checklist.md, config/image-config.yaml) and the body stays an overview. Not 5: no bundle files exist to verify these paths resolve, the resource list is duplicated (prose paragraph plus trailing bullets), and some inline bulk (host-specific attachment paths, detailed vendor strategy notes) belongs in a reference file.	4 / 5
	Total	17 / 20 Passed

Description

75%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.

A specific, third-person description that clearly answers both what (multi-vendor routing with named vendors, models, and auth mechanisms) and when (explicit 'Use for' clause with natural trigger terms). Consistently lands just below the top anchor due to non-exhaustive action and trigger coverage.

Dimension	Reasoning	Score
Specificity	"Multi-vendor AI image generation with authentication-aware parallel dispatch" and "Routes to Codex (gpt-image-2 via ChatGPT OAuth) and Pollinations (flux/zimage, free with signup)" name the domain plus several concrete capabilities with specific models and auth mechanisms. It stops short of 5 because the full action set (reference images, manifests, vendor comparison) is not enumerated.	4 / 5
Completeness	The 'what' is concrete (multi-vendor routing to named vendors and models) and an explicit "Use for..." clause answers 'when'. Not 5 because the 'when' is generic and omits concrete trigger scenarios such as reference images or vendor comparison.	4 / 5
Trigger Term Quality	"Use for image generation, image creation, visual asset generation, and AI art" gives good natural-phrase coverage users would actually say. Not 5 because common synonyms the body itself uses (illustration, concept art, product photos, mockup) are missing from the trigger list.	4 / 5
Distinctiveness Conflict Risk	Named vendors (Codex, Pollinations, Gemini) and auth-aware routing carve a distinct niche with low conflict risk. Not 5 because broad triggers like "image generation" and "AI art" could still collide with other image-generation skills.	4 / 5
	Total	16 / 20 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: first-fluke/oh-my-agent
Path: benchmarks/runs/oma/.agents/skills/oma-image/SKILL.md
Commit: ce19702

Reviewed: 4 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.