ccg
Claude-Codex-Gemini tri-model orchestration via /ask codex + /ask gemini, then Claude synthesizes results
40
40%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/ccg/SKILL.mdQuality
Discovery
17%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 reads more like a technical implementation note than a skill description meant to help Claude select the right skill. It lacks natural trigger terms users would use, provides no 'Use when...' guidance, and doesn't clearly explain what problems this skill solves or what outputs it produces.
Suggestions
Add an explicit 'Use when...' clause describing scenarios like 'Use when the user wants to compare answers from multiple AI models, needs a synthesized response combining different perspectives, or asks for multi-model analysis.'
Replace jargon like 'tri-model orchestration' with natural language describing the benefit, e.g., 'Queries both Codex and Gemini, then synthesizes their responses into a unified answer to leverage multiple AI perspectives.'
Include natural trigger terms users might say, such as 'compare models', 'get multiple opinions', 'cross-reference with other AIs', 'multi-model', or 'combined analysis'.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | It names a specific pattern ('tri-model orchestration') and mentions concrete actions (asking codex, asking gemini, Claude synthesizes), but the actual capabilities and what problems this solves are not clearly articulated beyond the orchestration mechanism itself. | 2 / 3 |
Completeness | The description only partially addresses 'what' (orchestrating three models) but completely lacks a 'when' clause or any explicit trigger guidance. There is no 'Use when...' or equivalent, which per the rubric caps completeness at 2, and the 'what' is also weak, so it scores 1. | 1 / 3 |
Trigger Term Quality | The description uses technical jargon like 'tri-model orchestration' and specific command syntax ('/ask codex', '/ask gemini') that users would not naturally say when requesting this capability. It lacks natural language trigger terms describing when or why someone would want multi-model synthesis. | 1 / 3 |
Distinctiveness Conflict Risk | The tri-model orchestration concept is somewhat distinctive, and the specific mention of Codex and Gemini narrows the scope. However, without clear use-case boundaries, it could overlap with other multi-model or comparison skills. | 2 / 3 |
Total | 6 / 12 Passed |
Implementation
62%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides a clear orchestration workflow with good fallback handling and explicit step sequencing. Its main weaknesses are the duplication between the 'How It Works' overview and the 'Execution Protocol' detail, and the lack of concrete examples showing what decomposed prompts and synthesized outputs actually look like. Adding a worked example would significantly improve actionability.
Suggestions
Add a concrete worked example showing a real request decomposed into specific Codex and Gemini prompts, plus a sample synthesized output with the agreed/conflicting/rationale/checklist structure.
Remove or collapse the 'How It Works' section into the 'Execution Protocol' section to eliminate duplication — the text diagram repeats what the detailed steps already cover.
Add a brief example of what the artifact files look like (even a 3-line snippet) so Claude knows what to expect when reading from `.omc/artifacts/ask/`.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient but includes some redundant explanations. The 'How It Works' section and 'Execution Protocol' section largely duplicate each other — the numbered text diagram repeats what the detailed protocol already covers. The 'When to Use' bullets are helpful but could be tighter. | 2 / 3 |
Actionability | Provides concrete CLI commands (`omc ask codex`, `omc ask gemini`) and artifact paths, but the decomposition step is abstract — there are no concrete examples of what a good Codex prompt vs Gemini prompt looks like for a real task. The invocation example is helpful but the synthesis output format lacks a concrete example. | 2 / 3 |
Workflow Clarity | The workflow is clearly sequenced (decompose → invoke → collect → synthesize) with explicit steps. Fallback handling covers single-provider and zero-provider scenarios. The synthesis step specifies what the output must contain (agreed, conflicting, rationale, checklist). The note about skill nesting limitations is a useful validation checkpoint. | 3 / 3 |
Progressive Disclosure | The content is reasonably well-structured with clear sections, but everything is inline in a single file with no references to supporting materials. The 'How It Works' overview followed by the detailed 'Execution Protocol' provides some layering, but the duplication between them hurts rather than helps. No bundle files are provided to offload detail. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- Yeachan-Heo/oh-my-claudecode
- Commit
679b418
- Reviewed
Table of Contents
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.