abpai
- 40 skills
- 0 followers
- 11 hours ago last updated
- ▌ Wildcard · abpai bundleHelp users interact with wildcard — a Bun-first CLI that captures file activity into a local timeline for search and summary. Covers running commands, configuring the tool, managing the macOS daemon, and troubleshooting.
- ▌ X CLI · abpai bundleUsage reference for the x CLI, a read-only X (Twitter) API client for the terminal built on Bun. Covers all commands, flags, authentication, and common usage patterns. Use when working with or answering questions about the x CLI.
- ▌ Code · abpai bundleRoute coding workflows through one scoped /code command. Use prepare-pr for effort-scaled PR readiness through push, simplify for behavior-preserving cleanup and high-signal test pruning of a target or a ranked whole-repo proposal, deslop-comments for comment cleanup that keeps rationale and directives, handoff for continuation prompts, and understand for a runnable real-code snippet, with an HTML code map on request.
- ▌
- ▌
- ▌ Claude · abpai bundleRun Claude Code as a headless worker — delegate, monitor, or resume Claude runs and named subagents, including background agents and a choice between CLI and Agent SDK execution. For local transcript inspection, use claude-session.
- ▌ Simplify · abpai bundleRoute scoped code simplification and whole-repository simplification proposals to the code workflow module.
- ▌ Cursor · abpai bundleRun, monitor, resume, or review work with Cursor Agent and Cursor models — bounded implementation, investigation, planning, or read-only review, with durable headless artifacts and exact chat continuation.
- ▌
- ▌ Doctor · abpai bundleRoute verification-first repo readiness and diff self-review audits to the doctor module.
- ▌ Prepare Pr · abpai bundleRoute effort-scaled PR-readiness work through validation, commit, seal, push, and PR update.
- ▌ Understand · abpai bundleRoute code-path understanding requests to the runnable real-code snippet module, with an optional HTML map.
- ▌ Distill · abpai bundleDecompose a complex system (codebase, architecture, paper, transcript, or document set) into essential primitives and a compressed mental model, without a line-by-line walkthrough. Prefer lateral-thinking for non-obvious hypotheses or cross-domain mechanism transfer.
- ▌
- ▌ Capture · abpai bundleRoute scoped behavior-capture (characterization safety net) work to the capture module.
- ▌ Harness · abpai bundlePrepare a repository for reliable agent-driven development. Adopt builds the behavior baseline, agent guidance, proof contract, and enforcement; doctor audits readiness or a diff; capture protects one uncertain behavior before it changes.
- ▌ Show Me · abpai bundleHelp the user understand the current topic visually with concise diagrams, code-shape sketches, and focused HTML artifacts. Use when the user asks to see, picture, sketch, or diagram how something works or what a change looks like.
- ▌
- ▌ Tutorial · abpai bundleWrite hands-on, self-verified, code-first tutorials where every step ends in a runnable action and the reader finishes able to do the thing.
- ▌
- ▌ Visualize · abpai bundleGenerate single-file HTML visualizations that teach systems, plans, code flows, architectures, comparisons, timelines, and concepts — for when the visual artifact itself is the deliverable.
- ▌
- ▌
- ▌
- ▌ Bun Expert · abpai bundleBun runtime guidance for JavaScript and TypeScript projects — package management, HTTP servers, Bun-specific APIs (`Bun.serve`, SQL, S3, Redis, `Bun.Image`, `Bun.WebView`, `Bun.$`), testing, bundling, and migrating from Node.js.
- ▌ Codex Exec · abpai bundleLaunch or monitor the Codex CLI as a headless worker for delegated implementation, review, or a second opinion, and to resume a session with new work. For a bare resume with no new task, or local transcript inspection, use codex-session.
- ▌ Antigravity · abpai bundleRun Google Antigravity CLI as a headless worker for bounded implementation, investigation, planning, or repository review, with Gemini model routing, durable liveness artifacts, and exact conversation continuation.
- ▌ CI Optimize · abpai bundleRoute measured CI cost and latency improvements through the engineering workflow pack.
- ▌ Engineering · abpai bundleRoute engineering-practice workflows through one scoped /engineering command. Use grill-me to stress-test a plan, tdd for behavior-first red/green/refactor, zoom-out for orientation, improve-architecture for deeper modules, defined-terms for DDD glossaries, complexity-report for read-only complexity findings, ci-optimize for measured CI improvements, and reduce for plan simplification.
- ▌ Defined Terms · abpai bundleBuild a DDD-style ubiquitous-language glossary from this conversation, flag ambiguities, and save it to DEFINED_TERMS.md.
- ▌ Human Writer · abpai bundleRewrite AI-sounding drafts into natural, specific prose — for text that feels generic, over-polished, promotional, repetitive, or too tidy.
- ▌ Codex Session · abpai bundleLocate, read, and summarize local Codex CLI transcripts by session or rollout UUID — bounded recent context plus opt-in tool summaries. Local-only: never launches Codex or contacts a model provider; use codex-exec to run or resume with new work.
- ▌ Hexagon Audit · abpai bundleAudit Ports & Adapters / hexagonal architecture boundaries in packages/ + adapters/ monorepos. Use for hexagon compliance, port/adapter separation, inward dependency flow, peer-adapter imports, and vendor SDK leaks.
- ▌ Status Update · abpai bundleSummarize a long-running agent task as a concise, evidence-backed snapshot — completed outcomes, current activity, remaining work, friction, surprises, and any required user action.
- ▌ Claude Session · abpai bundleLocate, read, and summarize local Claude Code transcripts by session UUID — bounded recent context, opt-in tool summaries, and child subagent transcript paths. Local-only: never runs Claude or contacts Anthropic; use claude to run or resume with new work.
- ▌ Complexity Report · abpai bundleProduce a read-only, evidence-ranked complexity and performance report.
- ▌ Improve Prompt · abpai bundleImprove, rewrite, or sharpen prompts into reusable instructions for planning, coding, review, eval, or decision-making — cutting vagueness, sycophancy, and success theater. Not for copyediting or humanizing prose.
- ▌ Improve Architecture · abpai bundleRoute architecture-improvement requests to the deep-module review module.
- ▌ Lateral Thinking · abpai bundleGenerate non-obvious hypotheses by decomposing a problem to its mechanism and raiding distant fields for transferable patterns, beyond standard advice. Use distill first when the system itself is still unclear.
- ▌ CLI Design Expert · abpai bundleDesign or review a command-line interface for usability and automation, backed by the clig.dev guidelines — help text, flags, output streams, exit codes, safety, and scriptability.