hashgraph-online/awesome-codex-plugins / plugins/beastlabai/multi-llm-plugin/skills/multi-llm
multi-llm
FlaggedOrchestrate plan reviews, task generation, implementation and code reviews across several LLM CLIs in parallel, or ask every model one free-text question about a plan. Use when the user says: review my plan, multi-llm, run a code review, generate tasks from a plan, implement this plan, ask the models about my plan.
schemas/review_suggestions.schema.json
View on GitHub{
"$schema": "http://json-schema.org/draft-07/schema#",
"title": "Plan / Task Review Suggestions",
"format_version": 1,
"description": "The JSON array `review-plan` and `review-tasks` ask each model for. Transcribed verbatim from the shipped prompt contracts (prompts/plan_review.txt and prompts/task_review.txt, whose 'Output Format' sections are byte-identical in field set, enum values and the 'exactly these fields' rule) — it is NOT an invented shape. One file serves both phases because the document is the same document; splitting it would create two copies free to drift. `format_version` is carried as a schema-level annotation rather than a document property because the document is an array and has nowhere in-band to put one (consolidated.schema.json, an object, carries its own as a property).",
"type": "array",
"items": {
"type": "object",
"required": ["title", "desc", "importance", "reference", "type"],
"properties": {
"title": {
"type": "string",
"description": "5-10 word summary of the issue"
},
"desc": {
"type": "string",
"description": "Text explaining the issue, in the form the phase's prompt asks for"
},
"importance": {
"type": "string",
"enum": ["high", "medium", "low"]
},
"reference": {
"type": "string",
"description": "Which section of the plan this affects (review-plan), or the task id / plan section (review-tasks)"
},
"type": {
"type": "string",
"enum": ["addition", "modification", "deletion", "clarification"]
},
"scope_relation": {
"description": "Optional: the reviewer's provisional scope classification of the finding's proposed action, \"in_scope\" | \"scope_extension\" | \"unclear\" (plans/scope-policy-and-report-tags.md §3). Deliberately untyped: a missing or malformed value reads \"unclear\" (utils/scope_policy.normalize_scope_fields) and never discards the finding. Validation returns the authoritative group classification; this one is a provisional report label."
},
"scope_reason": {
"description": "Optional: one short line saying why, against the scope baseline. A definite scope_relation without a reason reads \"unclear\"."
},
"scope_reference": {
"description": "Optional: the baseline section or text the classification rests on, or \"\"."
}
},
"additionalProperties": false
}
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirements | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |