Productivity

alibaba Bundle

Use this skill when writing regular expressions, debugging pattern matching,optimizing regex performance, or implementing text validation. Triggers on regex, regular expressions, pattern matching, lookahead, lookbehind, named groups, capture groups, backreferences, and any task requiring text pattern matching.

alibaba Bundle

Use this skill when writing bash or zsh scripts, parsing arguments, handling errors, or automating CLI workflows. Triggers on bash scripting, shell scripts, argument parsing, process substitution, here documents, signal trapping, exit codes, and any task requiring portable shell script development.

luko248 Skill

Charts, tables, calendar, pagination, and data helpers.

fieldlu Bundle

模型效果不好（或"好得可疑"）时的系统诊断流程。当用户说"模型精度低/loss 不降/泛化差/过拟合/ 欠拟合怎么办"、报告训练集表现与验证集脱节、或纠结该加数据还是加模型容量时激活。 核心动作: 先用偏差-方差分解(bias-variance decomposition)把问题归因到欠拟合(underfitting)/ 过拟合(overfitting)/数据问题三类，再对症下药；同时把"训练精度 100%"当红旗而非捷报 （完美可分可能恰是过拟合假象）。覆盖: 学习曲线判读、正则/剪枝强度取舍、调参后全量重训纪律。 不适用于: 还没选模型的选型阶段（用 ml-task-matching）、性能度量本身选错了的情况 （先确认尺子再诊断数值）、多算法科学比较（用算法比较检验流程）。

fieldlu Bundle

SVM 使用决策树：可分性判断→硬/软间隔→核函数选型→C 与 γ 调优→规模可行性。用户问 "SVM 参数怎么选""选哪个核 / kernel choice""C 和 gamma""支持向量机适用场景"，或完美可 分而沾沾自喜、大数据上硬跑非线性核时激活。纪律：默认软间隔(完美线性可分可能是过拟合假 象)；文本用线性核、不明先试高斯核；核选择是最大变数且未决；非线性核不低于 O(m²)。 不适用于：横向选型(ml-task-matching)、需自然概率输出的场景。trigger: support vector machine, kernel trick, soft margin, hinge loss, RBF kernel, 核函数。

fieldlu Bundle

选机器学习算法/模型时的第一性原理检查器。当用户在问"哪个算法/模型最强"、"该用什么模型做 X"、 想直接照搬排行榜(benchmark/leaderboard)结论、或团队为选型争论不下时激活。 核心动作: 先刻画任务本身(样本量/噪声/数据结构), 再匹配归纳偏好(inductive bias / inductive preference), 而非寻找"全面最优"算法——NFL 定理(no free lunch)证明脱离具体问题谈优劣无意义。 也适用于: 两个模型打平时的决胜依据、评测榜单结论能否外推到自己任务的判断。 不适用于: 已选定模型后的效果诊断(用 ml-diagnosis)、纯科普问答"什么是 NFL 定理"。

fieldlu Bundle

神经网络训练决策纪律：何时用 NN、结构怎么定、训不动怎么救。用户说"神经网络不收敛/loss 不下降""该几个隐藏层""learning rate 怎么调""梯度消失"，或训练振荡、不同种子结果差异巨 大、纠结早停时机时激活。动作：判可分性定结构下限(非线性可分必须加隐层)→按 loss 曲线分 诊病因→多组初值/模拟退火/SGD 跳局部极小(无理论保障)→早停+正则化两手。不适用于：横向选 型(ml-task-matching)、训练正常但泛化差的归因(ml-diagnosis)。trigger: neural network not training, backpropagation, hidden layer, local minimum, early stopping, 梯度消失。

fieldlu Bundle

优化方法选型：拿到优化问题先问"凸不凸/有无约束/梯度贵不贵"。用户问"牛顿法还是梯度下降/ L-BFGS 用在哪/KKT 有什么用/SVM 为何转对偶"、纠结学习率收敛诊断、担心陷局部最优时激活。 动作: 凸性判定→一阶vs二阶→拉格朗日与KKT对偶→收敛曲线归因→高维鞍点重估。 不适用于: 训练排障(ml-deep-training-playbook)、横向选型(ml-task-matching)。 trigger: convex optimization 凸优化, L-BFGS, KKT conditions, duality 对偶, 鞍点

garrettjsmith Skill

When the user wants to create, schedule, or optimize Google Business Profile posts. Also use when the user mentions "GBP posts," "Google posts," "GMB updates," "business profile posts," "what should I post on Google," or "GBP content calendar." For overall GBP optimization, see gbp-optimization.

guiziweb Skill

Add custom logic on Create/Update/Delete operations — send an email, dispatch a command, publish an event, push to a queue.

lukaskellerstein Bundle

Design the complete work hierarchy for a Paperclip company — Goals, Projects, and Tasks as a unified top-down planning workflow. Produces ._planning.json for deterministic generation of goals/, projects/, and tasks/ directories. Use when creating a new company's work plan, reorganizing work structure, or linking goals to projects and tasks.

prowlrbot Bundle

Use this skill when the user wants to automate git operations, manage branches, generate commit messages, handle PRs, set up git hooks, or implement a git workflow.

prowlrbot Bundle

Wake up to a personalized daily summary covering your calendar, unread emails, local weather, and top news stories.

prowlrbot Bundle

Use this skill when the user wants to draft, improve, or respond to emails — cold outreach, follow-ups, escalations, team announcements, or any professional communication.

prowlrbot Bundle

Use this skill when the user has a meeting recording, transcript, or notes and wants a structured summary with action items, decisions, and follow-ups.

prowlrbot Bundle

Use this skill when you need Claude to return structured JSON — data extraction, classification, form parsing, or any task needing machine-readable output.

alihusains Skill

Maps a real process into a swimlane by role, capturing every handoff, decision, wait, and rework loop, distinguishing the process as performed from the process as designed, and quantifying where time is actually lost. Use when a cross-team process is slow, error-prone, or disputed, or before documenting or automating it; trigger on 'map this process', 'draw the workflow', 'swimlane diagram', 'where does this get stuck', 'nobody agrees who does what here'. Not for writing the step-by-step instructions once the flow is agreed (use operations-sop-authoring), not for system architecture diagrams (use engineering-decision-record), and not for a product feature flow (use product-requirements-doc).

bxmaximum Skill

Covers business processes (bizproc) — start/stop workflows from PHP, document types, custom activities in /local/activities/, boundary vs agents/Messenger; mentions bizprocdesigner. Applied for approvals, document workflows, automation activities. Key terms — CBPDocument, CBPRuntime, CBPWorkflow, CBPActivity, document type, Starter, workflow template, activities.

bxmaximum Skill

Covers input data validation in Bitrix — ValidationService (main.validation.service), attributes #[NotEmpty], #[Email], #[Length], #[Range], #[RegExp], #[InArray], Request DTO via ValidationParameter autowire, rule attributes on controller action parameters, custom validators via AbstractPropertyValidationAttribute + ValidatorInterface, aggregation of errors in ErrorCollection. Applied when checking input of controllers, services and CLI commands, validation of forms, DTO and action method parameters. Key terms — ValidationService, NotEmpty, Email, Length, ValidationParameter, Request DTO, validator, constraint.

dragoon0x Skill

Task sequences, progressive disclosure, and the pacing of an experience over time. Use when designing multi-step flows, evaluating onboarding, or diagnosing why users get lost.

srobinson Skill

Spawn N parallel candidates at the same task, pick a base, graft the strongest parts of the losers into it. Use for /arena, 'arena this', 'throw it in the arena', or when one attempt at a non-trivial artifact would lock in the wrong shape.

srobinson Skill

Design an auditable playbook when no narrower one fits: a large migration, an ambitious multi-part change, or work a human reviews after stepping away. Scales rigor to the task, runs a hypothesis loop, and logs decisions via show-me-your-work. Use for /figure-it-out, 'figure it out', a large migration, or when no narrower playbook applies.

omicverse Bundle

Run OmicVerse single-cell differential abundance or compositional analysis as a reusable, triggerable skill. Use when comparing cell-type abundance between conditions in AnnData, choosing between scCODA, milopy, and milo backends, or adapting a related notebook into a repeatable DCT workflow.

omicverse Bundle

Run OmicVerse single-cell differential expression analysis as a reusable, triggerable skill. Use when comparing conditions inside one or more cell types in AnnData, choosing between Wilcoxon, t-test, and memento backends, or adapting a related notebook into a repeatable DEG workflow.

harshanandak Bundle

Toggle Forge's workflow gates and rails — strong-but-toggleable enforcement. `forge gate enable|disable <gate-id>` flips `workflow.gates.<id>.enabled` in `.forge/config.yaml`; the installed git hooks read resolved config at run time, so disabling a rail makes them genuinely inert. `forge gate approve|reject <issue> <gate>` records durable human-gate approval events; `forge gate status`/`check` query them; `forge control <id> <mandatory|optional|permission>` sets tri-state DECLARED intent (writes the same `enabled` field — no independent runtime enforcement). Use when the user says "disable the gate", "turn off TDD enforcement", "the tdd intent rail is blocking me", "toggle or enable a gate", "loosen enforcement", or "approve a human gate". Common default-ON toggleable rails: `rail.tdd_intent`, `rail.kernel_tracking`, `rail.auto_shepherd` — e.g. `forge gate disable rail.tdd_intent`. NOT for addressing PR review feedback (review), NOT the status snapshot of work in flight (status).

harshanandak Bundle

The flagship Forge orchestrator: given a goal or a ready issue, it composes the stage skills (triage-ready · claim-safety · plan · dev · validate · ship · review · verify) into the right path for the work, running autonomously between human gates and pausing at them. Use this whenever the user wants to work the next ready issue, drive a feature or fix end-to-end, "take this through to a PR", orchestrate the whole workflow with human checkpoints, or asks some form of "what should I work on and get it done" — even if they never say "smith" or "orchestrate". Reach for it especially when the request spans multiple stages (plan → build → ship) or asks to keep a human in the loop at intent, plan, or merge. Prefer a single stage skill only when the user explicitly wants just that one step (e.g. "just open the PR").

harshanandak Bundle

Forge kernel — umbrella index/router for a Forge project. Reach for this FIRST when orienting rather than executing: at session start, when you need the map of how the system fits together, or when unsure WHICH skill or `forge` verb a task belongs to. It indexes `smith` (end-to-end orchestrator), the stage ladder (plan → dev → validate → ship → review → verify), the utility/issue skills, and the day-to-day `forge` CLI verbs. Trigger on "how does the Forge workflow work", "which forge command or skill for X", "I'm new here, how is Forge set up", or "should this be a Forge issue or a TodoWrite". Index layer only — hand off the doing: rank/pick the next ready issue → `triage-ready`; current stage / "where am I" / active or stale work → `status`; create/update/close/search one issue → `issue-basics`; claim-then-prove ownership before mutating → `claim-safety`; drive one issue from plan to a merged PR under gates → `smith`; token-bounded state for the Hermes harness → `hermes-forge`.

harshanandak Bundle

Forge dispatch skill: the session bootstrap that makes Forge's skills auto-trigger as a reasoning-driven SYSTEM. Reach for this FIRST, before ANY response (including clarifying questions, codebase exploration, or file reads), whenever there is even a 1% chance a Forge skill applies. Carries the 1%-rule, the announce-before-acting rule, an anti-rationalization red-flags table, a subagent escape hatch, and an intent-to-skill routing table: add/build/scope a feature to plan; implement an existing task to dev; fix a failing test or bug to dev (debug first); type-check/lint/tests to validate; push and open a PR to ship; address PR/review feedback to review; post-merge health to verify; what to work on to triage-ready; where am I / current stage to status; create/close/search an issue to issue-basics; drive one issue end-to-end to smith; monitor a PR to shepherd. Run `forge skill for "<situation>"` for a deterministic fallback. NOT itself a stage; the kernel skill is the fuller umbrella index once oriented.

harshanandak Bundle

Surfaces, ranks, and EXPLAINS the single best next issue in a Forge project — strictly read-only. Recomputes the live ready queue and explains what is blocked via `forge issue ready`/`blocked`/`stats`/`show`, always against the live kernel, never `forge board`. Recommends ONE issue with a one-line reason, then hands off; never claims, comments, closes, or mutates. Use when the user asks "what should I work on", "what's next", "what's ready", "pick/triage my next task", "top of the ready queue", "which issue should I start", "why is this issue blocked", or "what's blocking the most work". NOT for claiming or taking ownership of the pick (use claim-safety), not for everyday issue create/update/list/search/close/comment/dep CRUD (use issue-basics), not for reporting the current workflow stage or "where am I / what's in flight" (use status), not for orienting a whole session or routing to stage skills (use kernel), not for driving an issue to a merged PR (use smith).

harshanandak Skill

Forge's front door — invoke FIRST when starting work in a project that has Forge available but you are not yet oriented: no `.forge/` yet, a fresh session, or the user says "/activation", "set up Forge", "start Forge", "is Forge active here?". Orients on the actual repo (detects project type, states findings) BEFORE offering anything, then offers the bare-minimum gates-disabled setup WITHOUT forcing it. Do NOT use this to run a specific stage (use /plan, /dev, /status, /ship directly) or once `.forge/` is already initialized and you know the workflow.

harshanandak Skill

Use before changing AGENTS.md — a workflow rule, a stage table, a glossary entry, or a maintainer note. Do not use for CLAUDE.md (it is a pointer and holds no content) and do not use for the generated harness mirrors under .claude/, .codex/, .cursor/, or .hermes/.

gtrabanco Bundle

Internal agentic-workflow maintenance: after SKILL.md edits, bump semver, lint authoring rules, and synchronize changelogs, READMEs, routing metadata, and migrations. Triggers: "bump the skill", "update the changelog", "version bump".

aws-samples Skill

Run the AI-DLC `incident-response` stage (operation phase) in isolation, without advancing the main workflow. Packages `/aidlc --stage incident-response --single`: the engine emits one run-stage directive for incident-response and its gate, the conductor runs it, then the single-stage run commits a synthetic-id pair and stops. The main workflow's Current Stage is never touched.

aws-samples Skill

Run the AI-DLC `application-design` stage (inception phase) in isolation, without advancing the main workflow. Packages `/aidlc --stage application-design --single`: the engine emits one run-stage directive for application-design and its gate, the conductor runs it, then the single-stage run commits a synthetic-id pair and stops. The main workflow's Current Stage is never touched.

aws-samples Skill

Run the AI-DLC `observability-setup` stage (operation phase) in isolation, without advancing the main workflow. Packages `/aidlc --stage observability-setup --single`: the engine emits one run-stage directive for observability-setup and its gate, the conductor runs it, then the single-stage run commits a synthetic-id pair and stops. The main workflow's Current Stage is never touched.

aws-samples Skill

Run the AI-DLC `reverse-engineering` stage (inception phase) in isolation, without advancing the main workflow. Packages `/aidlc --stage reverse-engineering --single`: the engine emits one run-stage directive for reverse-engineering and its gate, the conductor runs it, then the single-stage run commits a synthetic-id pair and stops. The main workflow's Current Stage is never touched.