Coding & Dev Tools

jie-meng Bundle

Create GitHub Pull Requests via `gh` CLI. Trigger when user asks to create, submit, or open a PR/pull request, or says phrases like "create PR", "提PR", "创建PR", "开PR", "提交PR", "submit PR", "open PR", "gh pr create", "发PR", "create pull request", "帮我提个PR", "PR提一下", "push and create PR". ZERO-SPECULATION RULE: Do NOT analyze the hostname — git.x.com, code.z.au are all potentially GitHub Enterprise. Just run `gh` and let it succeed or fail. Supports PR templates, auto-detecting the main branch, and filling template content from code changes.

jie-meng Bundle

Archive a completed work item by moving its directory into the archive with a date and type prefix. Marks the work item as done: the lifecycle state of a work item is its directory location — under changes/type/ it is active, under changes/archive/YYYY-MM-DD-type-name/ it is archived. Trigger: "fullstack archive", "archive it", "归档", "做完了", "ship it", "结了", "merged", "已完成", "收尾". Use when the user says a feature, fix, or refactor is done, shipped, merged, or finished and should be moved out of the active listing.

jie-meng Bundle

Propose a new work item across a multi-repo fullstack workspace — gather context, design the approach, validate unknowns (spike) if needed, and write the work-tracking documents in one step. Standard mode for clear requirements; deep mode (spike) for uncertain ones. Trigger: "fullstack propose", "fullstack plan", "fullstack spike", "全栈提案", "全栈规划", "全栈计划", "全栈 spike", "全栈探针", "全栈验证", "全栈 poc"; ALSO when the user asks to plan, design, or prototype a feature/fix/refactor before implementing — "plan this", "设计一下", "怎么做X", "先验证X", "we should add X". Planning only — does not edit project code or create branches.

jie-meng Bundle

Context-aware code review for git STAGED changes (git diff --cached). Triggers when user mentions "暂存" (staged) with any review verb: 审查/评审/看/检查. Examples: '审查暂存', '评审暂存', '看看暂存', '看一下暂存', '检查暂存', '暂存区审查', '看一下暂存代码', '看看暂存的代码', '审查暂存代码', '检查暂存代码'. English triggers: 'review staged', 'staged code review', 'check staged', 'look at staged', 'look at staged code', 'review staged code'. Reads related files to validate changes in context. Auto-generates commit message. ONLY reviews staged changes, NOT unstaged or all changes.

jie-meng Bundle

MANDATORY TRIGGER: Any URL containing `/pull/` + review intent = invoke this skill IMMEDIATELY. Do NOT analyze the hostname — git.x.com, gitlab.y.com, code.z.au are all potentially GitHub Enterprise. NEVER speculate about the platform; just run the skill and let `gh` succeed or fail. Zero pre-analysis, zero hesitation. Comprehensive PR code review via GitHub CLI (`gh`). Trigger on review/审查/评审/CR + PR URL or PR number. For routine PR actions (not full code review), prefer `gh-operations`.

mardab96 Skill

Checks whether cold emails are actually reaching inboxes by reading SPF, DKIM and DMARC records, sending volume, bounce and complaint rates. Use when reply rates fall without the copy changing, when a new domain starts sending, or when someone asks whether their emails are going to spam.

mardab96 Skill

Finds fields, steps and UX patterns that kill form completion. Use when traffic reaches the form and stops there, or before adding another qualifying field.

mardab96 Skill

Reviews a proposed or existing B2B lead magnet for buyer-stage fit, pain relevance, qualification strength and follow-up value. Use when you are about to build a new lead magnet, or when downloads are high and nothing moves after them.

mardab96 Skill

Works out why a prospect went quiet after a good call and what to send now, by reading the actual thread and call notes rather than assuming they got busy. Use when a promising deal stops replying, or when several deals go silent at the same stage.

mardab96 Skill

Learns from won and lost deals to fix upstream targeting and messaging. Use when a run of losses feels similar, or when you keep reaching proposal stage with people who never buy.

mardab96 Skill

Verifies that ads, landing page and CTA promise the same thing. Use when click-through looks fine but the page converts badly, or when leads arrive confused about what was offered.

mardab96 Skill

SDR Handoff Brief Generator

mardab96 Skill

Summarizes facts, hypotheses, next actions and approvals for the week. Use when a reporting week ends, or when someone needs one page on what happened and what needs approval.

mardab96 Skill

Reverse-engineers competitor angles you are losing to. Use when a competitor keeps winning the same deals, or when your angle stopped landing and nobody can say what changed.

ruicore Bundle

Evidence-led diagnosis for hard bugs and performance regressions, using the full reproduce → minimise → hypothesise → instrument → test → fix → prevent loop when possible and a bounded, non-mutating assessment when reproduction is unsafe, unavailable, production-only, or outside the user's authority. Use when the user asks to diagnose or debug a failure or regression; do not broaden an assessment-only request into implementation.

ruicore Bundle

Use this skill when the user asks Codex to challenge, critique, review, pressure-test, or "grill" an idea, plan, architecture, implementation, PR, design doc, migration, rollout, or technical decision. Behave like a direct senior engineer reviewer focused on production realism, code-backed evidence, scope control, risks, trade-offs, and concrete next actions. Do not use it for simple factual questions, ordinary code explanation, or tasks where the user clearly wants direct implementation instead of critique.

ruicore Bundle

Analyze the multi-track ai-signal-radar repository to produce monthly or quarterly reviews, theme and trend synthesis, cross-period comparisons, or idea-mining reports from existing Ray-reviewed data. Use for historical analysis of an explicit systems or creation track, or an explicitly requested cross-track comparison, based on track-scoped data, radars, themes, generated indexes, and monthly reports without ingesting new weekly reports.

ruicore Bundle

Draft, format, and safely publish Medium stories through an authenticated browser. Use for Medium article preparation, editor formatting, draft creation, or publication; do not use for generic blog writing with no Medium delivery.

ruicore Bundle

Draft, format, attach media, save, verify, and safely publish LinkedIn posts through an authenticated browser. Use for LinkedIn post preparation, draft editing, media upload, publication, or troubleshooting the current Shadow DOM composer; do not use for generic social copy with no LinkedIn delivery.

ruicore Bundle

Create polished Excalidraw diagrams from text descriptions. Use when the user asks for a diagram, flowchart, architecture diagram, decision tree, process flow, data flow, system overview, or similar visual explanation, or says "draw this", "diagram this", "visualize this", or "make a chart". Produces styled `.excalidraw` files with consistent colors and can export a shareable excalidraw.com URL.

ruicore Bundle

Refactor a working repository from broad, weakly owned modules into concept-owned packages and localized change surfaces while preserving behavior. Use when large or flat modules contain multiple independent responsibilities, agents repeatedly edit the same hotspots, ownership is ambiguous, or the user asks to redesign and implement the source or test layout. Do not use for LOC-only cleanup, formatting, a narrow bug, a greenfield package layout, or a read-only architecture review.

autonomous-ai Bundle

Explain how to run and use this repository's local Web UI, with detailed guidance for every Settings section and its visible controls. Use for questions about web development, Monitor, Setup, or Settings.

autonomous-ai Bundle

Tracks the USER's mood only — signals + synthesized decision from camera/voice/telegram. Do NOT use for emotion commands directed at the device ("show sad", "be happy", bare "sad now"); those go through emotion/SKILL.md and are never logged here. Music/wellbeing skills consume the latest decision.

autonomous-ai Bundle

Low-level speaker and microphone hardware control — adjust volume, play test tones, record raw audio. Do NOT use for TTS/speech (that is the Voice skill).

autonomous-ai Bundle

Tracks and analyzes behavioral patterns (habits) for known users based on their wellbeing, presence, posture, and activity history. Use when answering questions about a user's routines ("What are Leo's habits?", "Has Leo been keeping to his routine?", "Notice anything about my patterns?"), or when invoked from wellbeing/SKILL.md or posture/SKILL.md on a threshold nudge to refresh patterns and provide habit-aware phrasing. Also feeds music-suggestion/SKILL.md with `music_patterns` for personalized genre. Habit does NOT fire its own standalone nudge — it enriches wellbeing/posture threshold-nudge phrasing.

autonomous-ai Bundle

Play and stop music from YouTube through the device's speaker on user request.

autonomous-ai Bundle

Activate predefined lighting scene presets (reading, focus, relax, movie, night, energize) when the user asks for activity-based or environment lighting. Scenes control both color temperature AND brightness. Do NOT use for specific colors (use LED Control) or emotion expression (use Emotion).

speciliam Skill

Claude-conducted, Codex-executed version of the LinkedIn early-career weekly drain. Claude (Opus 4.8) is the parent/conductor — it builds the durable run state, then drains last-week Entry-level software-engineer results one stage at a time by launching a single fresh Codex worker per stage (reusing linkedin-early-career-weekly's run_stages.py with --max-stages 1), runs a reasoning checkpoint after every discover/tailor/apply stage to catch bad dedupe, wrong-resume drift, tracker corruption, and systemic browser failures, and reconciles the tracker at the end. Each Codex worker owns exactly one stage for one posting, then exits. Use when the user says `$linkedin-early-career-weekly-claude`, "conduct the early-career weekly with Codex workers", or wants the weekly drain run with Claude judgment between every stage instead of the blind run_monitored.py loop.

hugobowne Bundle

Use this skill any time you are opening a PR to ship code or the user tells you to "ship it". It explains how to write PR bodies effectively and follow the style guide.

hugobowne Bundle

Imagine future bug post-mortems for the codebase. Identifies fragile code, implicit assumptions, and likely failure modes by writing realistic incident reports for bugs that haven't happened yet.

hugobowne Bundle

Work inside a running marimo notebook's kernel — execute code, create cells, and build a notebook as an artifact. Use when the user wants to start a marimo notebook or work in an active marimo session.

hugobowne Bundle

This skill is for interacting with Anki through AnkiConnect, and should be used whenever a user asks to interact with Anki, including to read or modify decks, notes, cards, models, media, or sync operations.

hugobowne Bundle

Draft maintainer replies to GitHub PRs and issues in the user's voice. Use when the user asks to reply, respond, comment on, or review a PR/issue, or when drafting feedback for contributors. Handles approvals, change requests, closures, and issue triage. Copies result to clipboard.

hugobowne Bundle

Perform mutation testing on the codebase. Introduces deliberate bugs one at a time, checks whether the test suite catches each one, and reports on test suite gaps. Optionally implements missing tests.

hugobowne Bundle

Plan new software projects by decomposing the goal into MVP tasks, researching frameworks and libraries with subagents, and preferring opinionated prebuilt solutions over bespoke complexity. Use when starting a new project, scoping an idea, choosing a stack, or producing an implementation plan for a single-user MVP.

tss-blue-ai-lab Skill

Systematic approach to finding and fixing bugs in any codebase. Use when: debugging errors, investigating unexpected behavior, fixing failing tests, or resolving production issues.