Agent skills for Codex CLI

This skill should be used when the user asks to "validate plugin structure", "review manifest files", "check frontmatter compliance"…

Asset pipeline optimization, compression, streaming, and resource management for efficient game development and delivery.

Find the actual cause of an admitted failure instead of a plausible one using a systematic empirical method grounded in Agans and Zeller…

Keep code alive when its dependencies misbehave — timeouts on every remote call, circuit breakers, bulkheads, bounded pools and queries…

Optimize AgentDB performance with quantization (4-32x memory reduction), HNSW indexing (150x faster search), caching, and batch operations.

Automate browser interactions for web testing, form filling, screenshots, and data extraction.

Reverse Engineering Skill

MCP server for X/Twitter interaction - profile analysis, thread tracking, and social graph exploration

Product manager that decomposes requirements into actionable tasks with priorities and dependencies.

Orchestrates an end-to-end product management workflow from feature teardown and prior-art discovery to requirement evaluation, PRD…

Full PM playbook — triage backlog, prioritize, assign, track, report, escalate.

Review local git changes from a PMBOK-based product management perspective.

Categorizes, extracts underlying needs from, and acts on incoming user feedback within a rapid 15-minute product management workflow.

Structure a PM's weekly review and planning session.

Tracks daily progress, identifies blockers, and enforces task escalation rules in structured 5-minute check-ins.

Facilitates structured 20-minute weekly sprint planning, goal decomposition, and task prioritization for engineering teams.

Use when managing Node.js dependencies with pnpm - install packages, configure monorepo workspaces, set up pnpm catalogs, resolve…

Keep pnpm current: run pnpm self-update/corepack prepare, align packageManager in package.json, and bump pnpm/action-setup + pinned pnpm…

Archive a done POC out of the active set but keep it as a parity/regression oracle, moving it under .pocs/archive/.

Create a new POC-driven feature file under .pocs/ from any source (a sentence, an issue, a document, a URL).

List every POC in .pocs/ with its phase and gap counts (split by [agent]/[user]).