Coding & Dev Tools

tomevault-io Bundle

測試生成技能

mm-arturo Skill

Persist verified session knowledge and current work before closing Codex, restarting the computer, switching projects, or ending substantial work. Use when the user says "wrap up", "save session", "prepare for restart", "checkpoint", "closing the shell", or similar. Includes a retrospective, durable-learning updates, handoff creation, validation, commit, and backup. Do not use for an ordinary commit request that does not require session persistence.

mm-arturo Skill

Reconstruct project state and reload durable lessons after a restart, a new Codex session, context loss, or a handoff. Use when the user says "resume", "continue", "where were we", "pick up where we left off", "load the handoff", or similar. Verifies the repository against the handoff and reloads applicable project, global, and skill knowledge before work continues.

tinyc0der Bundle

Refines raw ideas into sharp, actionable concepts through structured divergent and convergent thinking. Use when an idea is still vague, when you need to stress-test assumptions before committing to a plan, or when you want to expand options before converging on one. Triggers on "ideate", "refine this idea", or "stress-test my plan".

tinyc0der Skill

Extracts what the user actually wants instead of what they think they should want. Achieves this through one-question-at-a-time interview until ~95% confidence about the underlying intent. Use when an ask is underspecified ("build me X" without "for whom" or "why now"), when the user explicitly invokes ("interview me", "grill me", "are we sure?", "stress-test my thinking"), or when you catch yourself silently filling in ambiguous requirements before any plan, spec, or code exists.

0xebby Skill

Add an emergency stop that halts entry paths (deposits, new orders) while never trapping user exits (withdrawals, claims). Use when a protocol moves money and you need a response to a discovered bug.

0xebby Skill

EIP-712 signature verification

teqfocusadmin Skill

Compress Claude's output tokens by 65–75% by responding in terse caveman-style prose. Drops articles, filler, pleasantries, and hedging. Keeps full accuracy, named references, metrics, and brand names exact. Brain still big. Mouth small. USE THIS SKILL when the user says: "caveman mode", "/caveman", "talk like caveman", "be brief", "terse mode", "quick answer", "save tokens", or "short version". THREE MODES: lite (drop filler only) · full (default — fragments OK) · ultra (max compression). STOP with: "normal mode" or "stop caveman". CRITICAL RULE: caveman mode is for Claude↔team communication ONLY. Never apply caveman compression to final deliverables — emails, proposals, decks, one-pagers, SOWs, case studies, or any asset that goes to a client or prospect. Client-facing output always uses the full Teqfocus voice.

teqfocusadmin Skill

Analyze claims, posts, or articles against authoritative sources and return a verdict with confidence score for each claim to combat misinformation. Use this skill when: user shares a claim and asks "is this true?", user forwards a social media post and asks you to verify it, user mentions "fact check this" or "verify this information", user asks "did [person] really say this?" or "did [event] really happen?", user is about to share information and wants to verify accuracy first, or user asks "is there evidence for [claim]?". Performs multi-source verification with 3-tier source hierarchy (gov/academic > major news > social), cross-references 3+ independent sources, checks timestamps, calculates confidence score 0-100, and assigns verdict: TRUE / FALSE / MISLEADING / UNVERIFIABLE.

teqfocusadmin Skill

Act as a Y Combinator partner providing brutally honest, strategic feedback on business ideas. Challenge assumptions, ask critical questions, and identify fatal flaws before they become expensive mistakes. Use this skill when: user pitches a business idea and wants honest feedback, user asks "what do you think of this idea?", user mentions "validate my idea" or "tell me what's wrong with this", user is about to invest time/money in a venture and wants a reality check, user explicitly requests "YC office hours style feedback", or user says "be brutal" or "don't hold back". Evaluates ideas against 8 fatal startup flaws, calculates YC fundability score (0-100), assesses unicorn potential, challenges hidden assumptions, and provides strategic pivot suggestions.

teqfocusadmin Bundle

ABM microsite skill for Teqfocus — builds account-specific, personalized microsites deployed on the Teqfocus WordPress site at teqfocus.com/abm/[company-slug]. Use this skill whenever the user wants to create a personalized microsite, ABM landing page, account-specific page, or any web asset targeting a named account or company. Also trigger when the user mentions: ABM page, personalized page, microsite, account page, company-specific landing page, Tier 1/2/3 account content, or wants to build something for a specific named company. This skill must be used for ALL Teqfocus microsite builds — do not attempt without reading this skill first. Always use this skill even if the request seems simple like "build a microsite for [Company]".

tinyc0der Skill

Manages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to another. Use when migrating a database schema in production, such as renaming or dropping a column without downtime (expand/contract). Use when deciding whether to maintain or sunset existing code.

tinyc0der Skill

Guides systematic root-cause debugging. Use when tests fail, builds break, something that worked yesterday broke, behavior doesn't match expectations, or you encounter any unexpected error. Use when you need to figure out what broke and why — a systematic approach to finding and fixing the root cause rather than guessing.

tinyc0der Skill

Instruments code so production behavior is visible and diagnosable. Use when adding logging, metrics, tracing, or alerting. Use when shipping any feature that runs in production and you need evidence it works. Use when production issues are reported but you can't tell what happened from the available data.

tinyc0der Skill

Checks completed work before handoff. Use when verifying a change or preparing evidence for review.

tinyc0der Skill

Researches architecture options. Use when comparing designs for a new subsystem.

datnt-cell Bundle

Create and lay out Unity UGUI screens

datnt-cell Bundle

Build Unity UI Toolkit (UITK) UIs

jrudman25 Skill

Stress-test a plan, decision, or idea through structured questioning. Use only when the user explicitly asks to be grilled, challenged, or interviewed about their reasoning.

teqfocusadmin Skill

TeqTalk pre-production skill. Use after the episode framework is confirmed. Produces the complete one-page interview cheat sheet Jas has open during recording: opening warmup, full set of core questions mapped to the framework, bridge questions, follow-up probes, and a strong closer. Formatted for quick scanning during a live conversation. Runs a 4-iteration refinement loop.

teqfocusadmin Skill

TeqTalk short-form reel extraction skill. Use this skill whenever a TeqTalk episode transcript is provided and the user wants to extract reels, clips, short-form content, or viral moments for YouTube Shorts, Instagram Reels, TikTok, or LinkedIn. Also trigger when the user says "extract reels", "find clips", "make shorts", "pull viral moments", "clip this episode", or "create reels from this transcript." Every output reel must be under 30 seconds. This skill governs ALL TeqTalk reel and clip extraction — do not attempt without reading this skill first.

teqfocusadmin Skill

Generate 6 two-line hook variations for any Teqfocus topic, post, article, reel, or newsletter — before the body gets written. Use this skill whenever the user asks for hooks, opening lines, "how should this start", "give me hook options", "this opening is weak", or is beginning any LinkedIn post, reel script, or newsletter and hasn't locked the first line. The hook is chosen BEFORE drafting, never bolted on after.

teqfocusadmin Skill

Write the pinned first comment for a Teqfocus LinkedIn post — the engagement anchor that extends the post, carries the link, or seeds the discussion. Use this skill whenever the user asks for a pinned comment, first comment, "what should we comment on our own post", "link in comments", or is finalizing any LinkedIn post that has a link, a longer resource, or needs discussion seeded. Every post with an external link uses this skill — links live in the comment, never the post body.

teqfocusadmin Skill

TeqTalk back-catalogue activation skill. Use when a list of 5+ past episodes is provided. Mines the archive for underperforming episodes with strong content that can be unlocked with new titles and thumbnails. Produces: priority ranking of episodes to recut, new title and thumbnail direction for each, 2 new clip cuts per episode, and a relaunch schedule. Based on the principle that updating titles and thumbnails on old videos can move a 2,000-view episode to 40,000+ views. Runs a 4-iteration refinement loop on titles.

teqfocusadmin Skill

TeqTalk editing skill. Use immediately after recording is complete in Riverside. Produces the Riverside-specific export and first-cut brief: which tracks to keep, top-and-tail timestamps, dead air and filler sections to cut, and export settings for handoff to Adobe. Designed so Jas can action it herself without re-listening to the full recording. Runs a 4-iteration refinement loop.

teqfocusadmin Skill

TeqTalk production skill. Use when a full transcript is available. Produces: a trimmed transcript with timestamps preserved, 7 key moment flags with clip briefs, and sections marked for removal. Never edits the substance of what was said — only marks sections for cutting. Also identifies the single wow moment, the best pull quote, and the anchor insight for asset production. Runs a 4-iteration refinement loop.

teqfocusadmin Skill

Turn any topic into a ready-to-publish LinkedIn post using a named copywriting framework — PAS, AIDA, BAB, STAR, or SLAY — adapted to Teqfocus voice. Use this skill whenever the user gives a topic and wants a structured post fast, mentions any of these frameworks by name, asks "format this as a post", "structure this idea", or wants multiple structural takes on the same topic. Pairs with teqfocus-hook-generator (hook first) and teqfocus-post-scorer (score after).

teqfocusadmin Skill

TeqTalk pre-production skill. Use after the guest brief is complete. Produces the 3–4 questions that make this specific episode worth recording — not a generic interview guide, but the specific arc for this guest, this topic, this moment. Includes the episode's central tension, the narrative arc, and the one wow-moment question. Runs a 4-iteration refinement loop. Trigger when episode framework or interview arc is needed.

qingmou314-design Bundle

LibTV 官方 CLI（libtv）：在命令行里完整操作 / 运行 LibTV 画布。 凡是和 LibTV 画布 / 项目 / 节点 / 模型 / 素材相关的操作，一律通过 libtv CLI 完成， 不要自己捏造 HTTP 请求或绕到网页端步骤。本 skill 内即包含完整的 CLI 命令操作手册； 常见场景见 examples/，安装/更新见 scripts/install.md。

datnt-cell Skill

Advise on Unity async and lifecycle strategy

datnt-cell Skill

Create, read and analyze C# scripts

datnt-cell Skill

Edit material and shader properties across Built-in/URP/HDRP

datnt-cell Bundle

Source-anchored design rules for DOTween 1.3.015

datnt-cell Bundle

Source-anchored design rules for UniTask 2.5.10

teqfocusadmin Skill

TeqTalk production skill. Use when episode theme and guest are confirmed — ideally before recording so the right frame can be captured. Produces 3 thumbnail concept variants (Type A: Dynamic Two-Shot, Type B: Guest Close-Up, Type C: Split with Text), each with expression direction, text overlay (5 words max), background colour spec, and A/B test logic. Includes the mobile-size check and swap trigger protocol. Runs a 4-iteration refinement loop. Based on THUMBNAIL_FORMULA.md system.

teqfocusadmin Skill

Split a TeqTalk solo episode transcript into 2 or more standalone videos of 10–15 minutes each, based on the total transcript length. Use this skill whenever the user uploads a transcript file and wants to split it into multiple videos, parts, or episodes. Also trigger when the user says "split this into videos", "cut this transcript into parts", "make episodes from this", "divide this transcript", or uploads any transcript and asks how to split it. The number of videos is determined automatically from the transcript duration — not fixed at two. Each output video must be complete and self-contained. Content is taken ONLY from the uploaded transcript — nothing invented, nothing added from outside.