houstongolden

houstongolden/bigbounce

Skill Claude CodeCodex

Builds reusable Convex components with isolated tables and app-facing APIs. Use for new components, reusable backend modules, integrations, or component boundary work.

houstongolden/bigbounce

Skill Claude CodeCodex

Plans Convex schema and data migrations with widen-migrate-narrow and @convex-dev/migrations. Use for breaking schema changes, backfills, table reshaping, or zero-downtime rollouts.

houstongolden/bigbounce

Skill Claude CodeCodex

Audits Convex performance for reads, subscriptions, write contention, and function limits. Use for slow features, insights findings, OCC conflicts, or read amplification.

houstongolden/bigbounce

Skill Claude CodeCodex

Creates or adds Convex to an app. Use for new Convex projects, npm create convex@latest, frontend setup, env vars, or the first npx convex dev run.

houstongolden/bigbounce

Skill Claude CodeCodex

Sets up Convex auth, identity mapping, and access control. Use for login, auth providers, users tables, protected functions, or roles in a Convex app.

houstongolden/bigbounce

Skill Claude CodeCodex

Routes general Convex requests to the right project skill. Use when the user asks which Convex skill to use or gives an underspecified Convex app task.

houstongolden/bigbounce

Skill Claude CodeCodex

When any quantitative claim changes (fNL number, σ, anomaly count, MCMC sample count, redshift, etc), grep every .tex AND every HTML/TSX page in the repo for the old value, replace with new in one atomic commit. Zero stale instances allowed. Claims-table on index.html must match paper abstract.

houstongolden/bigbounce

Skill Claude CodeCodex

Mirror a compiled paper PDF to every hosting path: public/papers/ (legacy static site) and site/public/ (Next.js site). Both must serve the same PDF. If only one is updated, one of the two site surfaces shows the stale PDF.

houstongolden/bigbounce

Skill Claude CodeCodex

Bigbounce-specific specialization of /revision-tracker-update. Binds the schema to project-context/peer-reviews/REVISIONTRACKER.md and to the 6-paper layout (P1A, P1B, P2, P3, P4, P5). Same workflow as the global skill but tied to this repo's file paths.

houstongolden/bigbounce

Skill Claude CodeCodex

Dual sync of bigbounce HTML pages (index.html, paper.html, activity.html, data-explorer.html, explained.html, glossary.html, figures.html, etc.) AND the Next.js site/ subdirectory. Both must update in same commit as SSOT — never as follow-up. Site staleness ≥20 fires = red flag.

houstongolden/bigbounce

Skill Claude CodeCodex

Bump paper version atomically: \paperVersion + \paperTimestamp + \date{} in the .tex + version.json at repo root. Semver rules: patch=minor edits, minor=substantive section adds, major=restructure. All 4 items in same commit. pypdf-verify page 1 reflects new metadata.

houstongolden/bigbounce

Instructions file CodexOpenCode

AGENTS.md instructions for houstongolden/bigbounce, covering agents.md — bigbounce (codex / any-agent entry point), ground truth, the site and research tooling.

houstongolden/bigbounce

Instructions file

Claude Code instructions for houstongolden/bigbounce, covering claude.md — bigbounce, ops center, review-round site sync (standing, houston 2026-06-11), paper status — always check ssot first and critical research directive.