ohong

ohong/agent-skills

Plugin Claude Code

Plugin marketplace listing 12 plugins: audit-and-improve, capture-design, distill-taste, fast-wifi, fieldtheory-research.

ohong/agent-skills

Plugin Claude Code

Chat with a research-grounded model of Lulu Cheng Meservey's taste for feedback and advice.

ohong/agent-skills

Skill Claude CodeCodex

Chat with a research-grounded model of Lulu Cheng Meservey for candid communications advice or work review. Use for "/ask-lulu", "what would Lulu think of...", "run this through Lulu", "Lulu's take on...", or requests to review a memo, founder communication, company narrative, project manifesto, or announcement…

ohong/agent-skills

Plugin Claude Code

Evidence-based repository audit and prioritized improvement plan.

ohong/agent-skills

Skill Claude CodeCodex

Audit a repository and produce an evidence-backed health report with a prioritized improvement plan. Use only for an explicitly requested broad codebase audit, tech-debt assessment, or remediation roadmap.

ohong/agent-skills

Plugin Claude Code

Save reference visuals as a reusable design-system swipefile.

ohong/agent-skills

Skill Claude CodeCodex

Save reference visuals as a reusable design-system swipefile. Use when the user explicitly asks to capture or preserve an aesthetic for reuse.

ohong/agent-skills

Skill Claude CodeCodex

Hand implementation work to Codex workers once planning is done. Invoke when Claude (Fable) has a plan/spec ready and is about to start coding — Fable specs and reviews, Codex implements. Skip for small edits (<30 lines) or pure judgment tasks.

ohong/agent-skills

Plugin Claude Code

Distill a person's publicly shared taste into a standalone ask- advisor skill.

ohong/agent-skills

Skill Claude CodeCodex

Deep-research a named person's (or group of people's) public writing, interviews, and talks, then distill their taste into a standalone installable ask-person advisor skill. Use for "/distill-taste name", "make a skill from a person's taste", "I want feedback from a person's perspective", or refreshing an existing…

ohong/agent-skills

Skill Claude CodeCodex

Walk the user through code changes made by coding agents so they understand what changed, why decisions were made, what trade-offs were accepted, what alternatives were or could have been considered, and what risks or follow-up choices remain. Use when the user invokes /explain, asks to understand an agent-made diff…

ohong/agent-skills

Plugin Claude Code

Diagnose macOS Wi-Fi and internet problems with measured checks.

ohong/agent-skills

Skill Claude CodeCodex

Diagnose a slow macOS network with the bundled measurement script. Use when the user asks to troubleshoot current Wi-Fi or internet performance.

ohong/agent-skills

Plugin Claude Code

Search, export, and analyze local Field Theory X bookmarks.

ohong/agent-skills

Skill Claude CodeCodex

Search, sync, export, and analyze the user's local X bookmarks with the Field Theory CLI while preserving full post text and provenance. Use when the user mentions Field Theory, ft, X bookmarks, bookmark folders, remembered saved posts, chronological bookmark exports, or bookmark-grounded research. Do not use for live…

ohong/agent-skills

Plugin Claude Code

Structured execution framework for long-horizon coding tasks, built on Boyd's OODA loop. Decomposes projects into milestones with validation gates, fresh context per step, and file-based state that survives compaction.

ohong/agent-skills

Skill Claude CodeCodex

Resume a mission from the last completed milestone. Use after /clear, in a new session, or when context is getting long. Reads plan and progress from .mission/ files and picks up where you left off.

ohong/agent-skills

Skill Claude CodeCodex

Mid-flight mission plan adjustment. Use when the approach is wrong, requirements changed, a milestone needs splitting, or scope needs updating. Boyd's "destruction and creation" applied to the plan.

ohong/agent-skills

Skill Claude CodeCodex

Interactive mission planning — research the codebase, ask clarifying questions, decompose work into milestones with acceptance criteria, and save an approved plan. Start here before /mission:start.

ohong/agent-skills

Skill Claude CodeCodex

Begin executing an approved mission plan from milestone 1. Requires .mission/plan.md to exist. Runs milestones sequentially with validation gates between each. Use /mission:plan first.

ohong/agent-skills

Skill Claude CodeCodex

Show current mission progress — milestones completed, current milestone, remaining work, and any blockers. Pure observation, no execution.

ohong/agent-skills

Skill Claude CodeCodex

Run acceptance criteria for the current milestone (or all milestones). Reports pass/fail. Pure observation — does not advance or fix anything.

ohong/agent-skills

Plugin Claude Code

Run a structured brand or product naming and validation process.

ohong/agent-skills

Skill Claude CodeCodex

Run a structured brand or product naming and validation process. Use for explicit naming strategy or candidate evaluation, not ordinary code identifiers.