Agent Claude Code
Reviews a feature segment in the main Civitai Next.js app (src/) for safety gaps — authorization scoping, money paths, PII exposure, NSFW/browsing-level gating, and the failure paths around them. Use before calling a segment done, alongside civitai-reuse-review, civitai-perf-review, civitai-test-review and…
Agent Claude Code
Scores a feature segment in the main Civitai Next.js app (src/) against the intent doc for the work — did the PR do what was actually asked, without quietly narrowing, widening, or transforming it. Use before calling a segment done, alongside civitai-reuse-review, civitai-correctness-review, civitai-perf-review and…
Agent Claude Code
Reviews a feature segment in the main Civitai Next.js app (src/) for production performance — N+1 queries, unindexed scans, hot feed path regressions, cache stampedes, event-loop blocking, and client bundle weight. Cleared for read-only prod EXPLAIN via the postgres-query and clickhouse-query skills. Use before…
Agent Claude Code
Reviews a feature segment in the main Civitai Next.js app (src/) for code that was rebuilt when it already exists — a component in src/components/, a service function in src/server/services/, a hook, a cache, a selector, a tRPC procedure — and reports pre-existing duplicate services the diff touches. Use before…
Agent Claude Code
Reviews the tests in a feature segment of the main Civitai Next.js app (src/) for whether they would actually fail if the code broke — vacuous assertions, over-broad mocks, fakes that hang instead of failing, suites that collect zero tests, and races in browser tests. Use before calling a segment done, alongside…
Agent Claude Code
Reviews the comments in a diff against the repo's comment guideline (CLAUDE.md → Coding Standards → Comments) — deletes what-narration, change-log narration and reviewer-justification, keeps the non-obvious why, trims the keepers to the fewest words that carry the fact, and flags comments that are now factually FALSE.…
Agent Claude Code
Creates single-page HTML design mockups following Civitai's design system (Mantine v7 + Tailwind). Use proactively when asked to create UI mockups, wireframes, or design variations.
Agent Claude Code
Given a session's commits or diff, finds the docs that the change made wrong — stale paths and symbols, checklist boxes now done, decisions recorded as open that something already settled, and two docs that now contradict each other. Also cuts padding, judging fact density rather than length. Reports exact replacement…
Agent Claude Code
Checks a migration slice against the main app's API surface — whether a local implementation duplicates an existing endpoint, whether an endpoint's parameters are being under-used, and whether a capability filed as unported already exists as an endpoint action. Use on any Retool slice before calling it done, alongside…
Agent Claude Code
The fourth review for a Retool migration slice — walks the export query by query and asks whether each behaviour is present in the build and matches. The only review that can see a capability that was never written; the three code reviews pass cleanly over a faithful implementation of the wrong thing. Use on every…
Agent Claude Code
Audits the Retool parity checklist against what is actually built, using the export inventories and the Retool screenshots. Verifies both directions — items marked done that are not, and items marked open that already work. Use when asked to check migration coverage, or before calling the Retool migration complete.
Agent Claude Code
Reviews a feature segment in any SvelteKit app (apps/moderator, apps/auth, apps/creator-studio) for duplication and missing abstractions — what should be a shared component, helper, or service, and where it belongs. Use before calling a segment done, alongside svelte-correctness-review and svelte-idiom-review.
Agent Claude Code
Reviews a feature segment in any SvelteKit app (apps/moderator, apps/auth, apps/creator-studio) for correctness — logic, data shape, authorization scope, and failure paths. Use before calling a segment done, alongside svelte-idiom-review and svelte-abstraction-review.
Agent Claude Code
Reviews a feature segment in any SvelteKit app (apps/moderator, apps/auth, apps/creator-studio) for Svelte 5 idiom (runes, async, forms, keys) and the shared UI conventions (shadcn primitives, text-dark-2, panel styling). Use before calling a segment done, alongside svelte-correctness-review and…
Agent Claude Code
Given a defect just found or fixed in a SvelteKit app (apps/moderator, apps/auth, apps/creator-studio), finds every OTHER place the same shape exists. Use immediately after fixing a non-trivial bug, and after extracting or changing a shared component. Not a review of a segment — a sweep for siblings of one known…
Command Claude Code
Remedy high seq scans on a table.
Command Claude Code
Sets up VS Code mention highlighting configuration for markdown files. Creates or updates the necessary VS Code workspace files to highlight @dev, @ai, and other @ mentions.
Hook Claude Code
Runs before the agent uses a tool for Bash and PowerShell tool calls, executing check-writable.mjs via node. From civitai/civitai.
Hook Claude Code
Runs after a tool call finishes for Write and Edit tool calls, executing check-svelte-ts.mjs via node. From civitai/civitai.
Settings file Claude Code
Agent settings declaring 2 hook events (PreToolUse, PostToolUse).
Skill Claude CodeCodex
Add a new ecosystem and base model to basemodel.constants.ts. Use when onboarding a new model family from providers like Baidu, ByteDance, Google, etc. Handles ECO constants, BM constants, ecosystem record, family (creates new if needed), license (creates new if needed), and base model record. Optionally triggers…
Skill Claude CodeCodex
Wire an existing ecosystem into the generation system. Adds generation support to basemodel.constants.ts, creates graph and handler files, and wires them into the ecosystem discriminator, workflow config, and router. Use after add-ecosystem when you need the ecosystem to show up in the generation form. Always checks…
Skill Claude CodeCodex
Author a prompt-enhancement system prompt for a new ecosystem and register/update it on the orchestrator's prompt-analysis service. Use when onboarding a new ecosystem (e.g. happyhorse, a new Flux variant, a new Wan video version) and the user provides the ecosystem key plus a reference link, model card, or…
Skill Claude CodeCodex
Wire an existing ecosystem into the LoRA training system so it appears as a trainable base model in the training form. Adds the base-model entry, schema enums, orchestrator validation, feature flag, and per-ecosystem default params. Use when a model family (already in basemodel.constants.ts) needs training support …