Best agent workflow skills, page 155

A skill your agent uses when choosing between install, enable, disable, or delete for an MCP server; when an agent needs to discover or delegate to MCP servers; or when MCP tools fail with…

friday-platform/friday-studio

Suede-owned artist campaign rollout. An agent skill from JasonColapietro/suede-creator-skills.

JasonColapietro/suede-creator-skills

Suede-owned launch packaging and install verification. An agent skill from JasonColapietro/suede-creator-skills.

JasonColapietro/suede-creator-skills

Suede-affiliated recurring marketing workflow design for cadence, inputs, decision rules, checkpoints, failure states, and measurable outputs.

JasonColapietro/suede-creator-skills

当任务完成、解决困难问题、用户要求总结经验/记录项目理解/更新项目规范，或 spec-end/spec-update 需要从当前 Spec 文档判断是否沉淀经验、知识、SOP、工具记忆、长期项目规范时使用。本 Skill 只做识别 + 分类 + 权重判断 + 草稿 + 用户确认，不落盘写入；内容已经明确、只需写入经验/知识文件并更新索引时直接用 exp-write；程序记忆（SOP）落盘成新…

HHU3637kr/skills

记忆写入 Skill，将重大经验写入 spec/context/experience/ 或知识记忆写入 spec/context/knowledge/，并更新对应索引（不写 MEMORY.md）。触发场景：/exp-write type=experience、/exp-write type=knowledge 手动写入，exp-reflect…

HHU3637kr/skills

当项目首次接入 Spec 驱动开发 / R&K Flow，需要创建 AGENTS.md、.agents/rules/、 .agents/skills/、spec/ 目录、记忆系统和 HTML 报告资产时使用。

HHU3637kr/skills

Toggle a project-local concise-output directive that suppresses agent prose padding during SDD steps.

haru/redmine_ai_helper

Show token usage for every SDD artifact in the active feature, the projected context window for each upcoming phase, and the savings already realized by token-budget (full backups vs current files).

haru/redmine_ai_helper

Provides rules for writing or updating a SKILL.md in this repository: frontmatter validation for license, name and description, the WHEN TO USE and WHEN NOT TO USE description format, and the…

IgniteUI/igniteui-webcomponents

[OMX] Strict autonomous loop: $deep-interview - $ralplan - $ultragoal (+ $team if needed) - $code-review - $ultraqa

yangyuan-zhen/PolyWeather

Guide for writing fullsend augmentation skills and sub-agents that work alongside shipped defaults.

fullsend-ai/fullsend

TRIGGER when: user asks to implement, fix, build, or work on something — whether from a docs/feat/wip plan OR a standalone task (bug fix, GitHub issue, one-off change).

serpro69/claude-toolbox

Sibyl 跨学科者 agent - 从认知科学、物理、生物等领域引入类比和方法. An agent skill from Sibyl-Research-Team/AutoResearch-SibylSystem.

Sibyl-Research-Team/AutoResearch-SibylSystem

Persistent memory system for Claude Code. An agent skill from rohitg00/awesome-claude-code-toolkit.

rohitg00/awesome-claude-code-toolkit

Intake protocol — the craft of turning a vague brief into an approved production plan.

alecs5am/ralphy

Launch bounded private Manus research, including Cobbler-managed Wide Research and deterministic reference fan-out.

aigorahub/elves

Compatibility alias for Cobbler external-agent setup. An agent skill from aigorahub/elves.

aigorahub/elves

Spawn and collect the reviewer fleet at stage20spawnreviewers.

closedloop-ai/claude-plugins

A skill your agent uses when my-pi-agent work needs requirement slicing, implementation planning, multi-step ordering, dependency decisions, or clear acceptance criteria before coding.

skuramatata/my-pi-agent

A skill your agent uses when my-pi-agent work needs requirement slicing, implementationPlan, DAG/order decisions, multi-step /code planning, or clear acceptance criteria before coding.

skuramatata/my-pi-agent

A skill your agent uses when Codex or OpenCode repeats avoidable mistakes, keeps guessing after failures, fails to trigger an expected skill/workflow/subagent, or the user corrects agent behavior…

skuramatata/my-pi-agent

A relentless interview that asks every frontier question at once, round by round.

DouglasNeuroInformatics/OpenDataCapture

当用户用 Claude 写小红书、公众号、视频脚本等内容时，输出出现翻译腔、英文思考、西式框架，导致“像外国人假装写中文”时使用。

majiayu000/spellbook

Polish the next file on the list to the house style without changing what it says, keep the original, and check it.

receptron/mulmoterminal

Measure the named documents with chaff and choose which to polish, within the agreed number — changing nothing yet.

receptron/mulmoterminal

Find what is worth tidying in this repository, measure each candidate's value and risk, and write a plan the person approves — including what to leave alone and why.

receptron/mulmoterminal

Read the named documents closely, address every structure problem chaff reports, find contradictions and ambiguities, and quote the text for each finding — changing nothing yet.

receptron/mulmoterminal

Measure the models with chaff, then write the style twice: chaff.yaml for what a machine checks, STYLE.md for what the writer reads — with a reason for every decision.

receptron/mulmoterminal

Collect the texts the person wants as models into .blueprint/sources/ as Markdown, and record where each came from.

receptron/mulmoterminal

Summarize the named documents to the agreed length, one sentence at a time, each backed by quotations, citing every part or leaving it out with a reason — writing .blueprint/summary.json and…

receptron/mulmoterminal

Extract every dated event, time and amount from the named documents into .blueprint/facts.json, each with a quotation it was read from — deciding nothing yet.

receptron/mulmoterminal

Write the next part of the outline to the house style, check it with chaff, cite what came from the sources, and mark it done.

receptron/mulmoterminal

Claude Code artifact schemas: plugin.json, frontmatter, hooks, settings, tools.

xiaolai/nlpm

The 50 NL programming rules (R01-R50) for skills, agents, commands, hooks, prompts.

xiaolai/nlpm

Autonomous Goal-directed Iteration. An agent skill from OpenLAIR/dr-claw.

OpenLAIR/dr-claw

Multi-persona idea evaluation with quality gate. An agent skill from OpenLAIR/dr-claw.

OpenLAIR/dr-claw

Loads the evaluation instance, searches GitHub for related repositories, builds a dataset description, queries the Prepare Agent for reference codebases, and downloads arXiv paper sources.

OpenLAIR/dr-claw

当确定要修改 AGENTS.md、commands、项目 AI 规则、skill/references、治理脚本，或明确优化 AI 驱动开发体系时使用；普通规则讨论和一次性纠偏不提前触发。

Peiiii/nextclaw

当确定要修改 AGENTS.md、commands、项目 AI 规则、skill/references、治理脚本，或明确优化 AI 驱动开发体系时使用；普通规则讨论和一次性纠偏不提前触发。

Peiiii/nextclaw

一种基于本能（Instinct）的训练系统，通过钩子（Hooks）观察会话，创建带有置信度评分的原子化本能，并将其进化为技能（Skills）、命令（Commands）或智能体（Agents）。

xu-xiang/everything-claude-code-zh

使用 AgentShield 扫描 Claude Code 配置（.claude/ 目录）中的安全漏洞、配置错误和注入风险。检查 CLAUDE.md、settings.json、MCP 服务端、钩子（Hooks）和智能体（Agents）定义。

xu-xiang/everything-claude-code-zh

使用 AgentShield 扫描您的 Claude Code 配置（.claude/ 目录）是否存在安全漏洞、配置错误和注入风险。检查项包括 CLAUDE.md、settings.json、MCP 服务器、钩子（Hooks）以及智能体（Agent）定义。

xu-xiang/everything-claude-code-zh

建议在逻辑间隔进行手动上下文压缩（Context Compaction），以便在任务阶段中保留上下文，而不是依赖任意的自动压缩。

xu-xiang/everything-claude-code-zh

Этот скилл MUST быть вызван для оптимизации запросов 1С: временные таблицы, JOIN vs подзапросы, параметры виртуальных таблиц, СКД, индексы.

Arman-Kudaibergenov/1c-ai-development-kit

Этот скилл MUST быть вызван когда пользователь говорит "напиши тест", "протестируй модуль", "запусти unit-тесты" для 1С.

Arman-Kudaibergenov/1c-ai-development-kit

Этот скилл MUST быть вызван когда у пользователя есть идея или задача, требующая дизайна и плана перед реализацией.

Arman-Kudaibergenov/1c-ai-development-kit

Этот скилл MUST быть вызван когда нужен паттерн работы с подсистемами БСП/SSL: пользователи, файлы, печатные формы, фоновые задания, версионирование, email.

Arman-Kudaibergenov/1c-ai-development-kit