📄 工具详情 ⚙️ 安装教程 📚 使用教程
能力标签
🔌 MCP🤖 Agent🔄 工作流🌐 翻译🐳 Docker💻 CLI🔗 REST API📚 RAG🖼 视觉🧠 Claude
⚙️
Agent工作流

开源AI工作流

基于 Python · 无代码搭建完整 AI 自动化流程
英文名：gald3r
⭐ 17 Stars 🍴 3 Forks 💻 Python 📄 NOASSERTION 🏷 AI 7.5分
7.5AI 综合评分
workflowagent-frameworkaiai-agentsai-memoryaiderpython
⚙️ 配置说明
✦ AI Skill Hub 推荐

开源AI工作流 是 AI Skill Hub 本期精选Agent工作流之一。综合评分 7.5 分，整体质量较高。我们推荐使用将其纳入你的 AI 工具库，帮助提升工作效率。

📚 深度解析
开源AI工作流 是一套完整的 AI Agent 自动化工作流方案。随着 AI 能力的不断提升，基于 Agent 的自动化工作流正在成为提升个人和团队效率的核心方式。区别于传统的 RPA 自动化（模拟鼠标键盘操作），AI Agent 工作流通过理解任务意图、动态规划执行路径，能够处理更复杂的非结构化任务。

开源AI工作流 工作流的设计遵循"最小配置，最大复用"原则：核心逻辑已经封装好，用户只需配置自己的 API Key 和业务参数即可快速上手。工作流内置错误处理和重试机制，在网络波动或 API 限速等情况下仍能稳定运行，适合作为生产环境的自动化基础设施。

在实际部署时，建议先在测试环境中运行 3-5 次，验证各个环节的输出结果符合预期，再部署到生产环境。AI Skill Hub 评分 7.5 分，是同类 Agent 工作流中的精选推荐。
📋 工具概览

AI开发框架，任务管理，23个代理，116+技能，50+命令。支持多种语言的AI开发框架，提供任务管理、代理管理和技能管理等功能，适合于AI开发和研究人员。

开源AI工作流 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排，将复杂的多步骤任务拆解为清晰的自动化流程，实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成，适合构建数据处理管线、业务自动化和 AI 辅助决策系统。

GitHub Stars
⭐ 17
开发语言
Python
支持平台
Windows / macOS / Linux
维护状态
轻量级项目，按需更新
开源协议
NOASSERTION
AI 综合评分
7.5 分
工具类型
Agent工作流
Forks
3
📖 中文文档
以下内容由 AI Skill Hub 根据项目信息自动整理，如需查看完整原始文档请访问底部「原始来源」。

AI开发框架，任务管理，23个代理，116+技能，50+命令。支持多种语言的AI开发框架，提供任务管理、代理管理和技能管理等功能，适合于AI开发和研究人员。

开源AI工作流 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排，将复杂的多步骤任务拆解为清晰的自动化流程，实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成，适合构建数据处理管线、业务自动化和 AI 辅助决策系统。

📌 核心特色
  • 可视化 Agent 工作流编排，无需编写复杂代码
  • 支持多步骤自动化任务链，实现全流程无人值守
  • 与外部 API、数据库和第三方服务无缝集成
  • 内置错误处理与自动重试机制，保障稳定运行
  • 提供可复用的自动化模板，快速在同类场景部署
🎯 主要使用场景
  • 自动化日常重复性工作，将精力集中于创造性任务
  • 构建数据采集 → 处理 → 输出的完整自动化管线
  • 实现跨平台、跨系统的数据流转和业务协同
以下安装命令基于项目开发语言和类型自动生成，实际以官方 README 为准。
安装命令
# 方式一：pip 安装（推荐）
pip install gald3r

# 方式二：虚拟环境安装（推荐生产环境）
python -m venv .venv
source .venv/bin/activate  # Windows: .venv\Scripts\activate
pip install gald3r

# 方式三：从源码安装（获取最新功能）
git clone https://github.com/wrm3/gald3r
cd gald3r
pip install -e .

# 验证安装
python -c "import gald3r; print('安装成功')"
📋 安装步骤说明
  1. 访问 GitHub 仓库获取工作流文件
  2. 在对应平台（Dify / Flowise / Make 等）中找到「导入工作流」功能
  3. 上传工作流文件
  4. 按照提示配置必要的环境变量和 API Key
  5. 运行测试确认流程正常后投入使用
以下用法示例由 AI Skill Hub 整理，涵盖最常见的使用场景。
常用命令 / 代码示例
# 命令行使用
gald3r --help

# 基本用法
gald3r input_file -o output_file

# Python 代码中调用
import gald3r

# 示例
result = gald3r.process("input")
print(result)
以下配置示例基于典型使用场景生成，具体参数请参照官方文档调整。
配置示例
# gald3r 配置文件示例（config.yml）
app:
  name: "gald3r"
  debug: false
  log_level: "INFO"

# 运行时指定配置文件
gald3r --config config.yml

# 或通过环境变量配置
export GALD3R_API_KEY="your-key"
export GALD3R_OUTPUT_DIR="./output"
📑 README 深度解析 真实文档 完整度 74/100 查看 GitHub 原文 →
以下内容由系统直接从 GitHub README 解析整理，保留代码块、表格与列表结构。

简介

<p align="center"> <img src="logo/Gald3r_Logo_Big.jpg" alt="gald3r banner" width="800"> </p>

<p align="center"> <strong>Song magic for your codebase.</strong><br> Persistent memory, multi-repo orchestration, and adversarial quality gates for AI coding agents across every major IDE. </p>

<p align="center"> <a href="LICENSE"><img src="https://img.shields.io/badge/License-FSL--1.1--Apache-purple.svg" alt="License: FSL-1.1-Apache"></a> <a href="CHANGELOG.md"><img src="https://img.shields.io/badge/version-1.5.2-green.svg" alt="Version"></a> <a href="https://www.python.org"><img src="https://img.shields.io/badge/python-3.10+-blue.svg" alt="Python 3.10+"></a> <a href="https://github.com/wrm3/gald3r"><img src="https://img.shields.io/github/stars/wrm3/gald3r?style=social" alt="GitHub stars"></a> </p>

---

What's Included

How activation works: All agents, skills, commands, hooks, and rules in this repo are inert markdown/script files until you copy the relevant platform folder into your project. Dropping .cursor/ into a project activates the full gald3r surface for Cursor. Same for .claude/, .agent/, .codex/, .opencode/, and .copilot/. You pick which platforms you use — nothing runs without you copying the files first.
ComponentCountWhat it covers
**Agents**22Task manager, code reviewer, QA engineer, project planner, infrastructure, ideas, verifier, project initializer, PCAC coordinator, and more
**Core Skills**100Recon suite (docs/file/repo/url/yt), research suite (deep/review/apply), release management, platform skills (21 platforms), medic, tier-setup, PCAC (10 skills), tasks, bugs, plan, project, features, subsystems, vault, constraints, code review, git, crawl, learn, subsystem-graph, swot, verify-ladder, and more
**Addon: Skill Packs**142 skills / 20 packsVercel, HuggingFace, Cursor Team Kit, Superpowers, AI Media, AI Video Tools, Cloud Providers, Blockchain, 3D Graphics, Community, Content Creation, Startup Tools, Phantom Connect, Context7, Continual Learning, Create Plugin, Firecrawl, Infrastructure, User Skills, and more
**Addon: Personality Packs**8 themesSilicon Valley, Star Wars, Star Trek, BSG, Firefly, Hackers, Shoresy, gald3r
**Commands**149Full @g-* command surface — task management, bug management, feature pipeline, release management, recon, research, code quality, vault, multi-repo, ideas, constraints, subsystems, swarm, and maintenance
**Hooks**17Session start, agent complete, pre-commit, pre-push, PCAC inbox check, vault operations
**Rules**15Always-apply: documentation, git workflow, error reporting, task completion gates, TODO/stub lifecycle, bug discovery
**IDE Platforms**2116 installer-managed + 5 reference-skill platforms (see Quick Start for full list)

Key Features

Feature Pipeline (Staging → Shipped)

Ideas don't go straight to tasks. gald3r now provides a structured feature pipeline — a staging layer between idea capture and task creation. Features move through four phases with checkpoints at each transition.

``` @g-feat-new "Webhook retry with exponential backoff"

Creates .gald3r/features/feat-036_webhook_retry.md

Appends approach to the feature's Collected Approaches table

@g-recon-repo https://github.com/some/project

5-pass deep analysis: skeleton → module map → feature scan → deep dives → synthesis

Apply approved findings into .gald3r/features/ staging

```

Feature lifecycle:

PhaseDescription
stagingResearch phase — collecting approaches from external analysis, discussions, external repos
speccedFormal requirements written — acceptance criteria defined
committedActive tasks created in TASKS.md — being coded
shippedFully implemented and verified

Features only become tasks (and enter the TASKS.md backlog) when promoted via @g-feat-promote. This keeps the task backlog clean and intention-driven.

---

- YAML frontmatter (id, priority, subsystems, dependencies)

New Installation

Step 1 — Clone the template for your tier:

```powershell

Full (core framework + installable skill & personality packs)

git clone https://github.com/wrm3/gald3r_template_full.git

Install specific platforms into a target project

powershell -ExecutionPolicy Bypass -File setup_gald3r_project.ps1 -TargetPath "C:\MyProject" -Platforms cursor,claude

Updating an Existing Install

If you already have gald3r installed (any version), the installer handles migration automatically:

```powershell

Or from inside your already-installed project (regenerate platform dirs only):

powershell -ExecutionPolicy Bypass -File setup_gald3r_project.ps1 -Platform auto # auto-detect running IDE powershell -ExecutionPolicy Bypass -File setup_gald3r_project.ps1 -Platform all # regenerate all installed platforms powershell -ExecutionPolicy Bypass -File setup_gald3r_project.ps1 -Platform all -Clean # wipe and regenerate ```

The update mode never touches your README.md, LICENSE, or CHANGELOG.md, never overwrites your non-gald3r skills and agents, and safely updates all g- prefixed framework files.

---

Optional: Docker MCP Server

gald3r is fully functional without any server infrastructure — the file-first architecture means every feature works with plain files. For teams that want semantic search, session memory across machines, Oracle database access, or server-side crawling, the gald3r Docker MCP server adds 42 server-backed tools.

The Docker server is a separate companion component (not included in this template). See the gald3r Docker server repository for setup instructions.

CategoryTools
**Memory**memory_search, memory_capture_session, memory_context, memory_ingest_session, memory_sessions, memory_capture_insight, memory_search_combined, memory_setup_user
**Vault**vault_search, vault_search_all, vault_sync, vault_read, vault_list, vault_export_sessions
**Install**gald3r_install, gald3r_plan_reset, gald3r_health_report, gald3r_validate_task, gald3r_server_status
**Crawling**platform_docs_search, platform_crawl_trigger, platform_crawl_status, crawl_add_target, crawl_list_targets, check_crawl_freshness
**Oracle**oracle_query, oracle_execute
**MediaWiki**mediawiki_page, mediawiki_search
**Video**video_analyze, video_batch_process, video_extract_frames, video_extract_transcript, video_extract_metadata, video_get_playlist
**Utility**md_to_html, config_reload, get_service_url

---

Environment Variables (Docker only)

Only needed if you run the optional Docker MCP server. Copy .env.example to .env:

POSTGRES_DB=rag_knowledge
POSTGRES_USER=gald3r
POSTGRES_PASSWORD=your_password_here
OPENAI_API_KEY=your-key-here

---

---

Install a pack into the current project

.\skill_packs\infrastructure\install.ps1

Install into a specific project directory

.\skill_packs\infrastructure\install.ps1 -ProjectRoot "C:\my-other-project" ```

To uninstall: delete the skill directories listed in the pack's PACK.md.

---

---

🤖 ai-media - AI video generation via Seedance 2.0 and Higgsfield cinematic models

SkillWhat it does
skl-seedanceSeedance 2.0 text-to-video and image-to-video via fal.ai, VolcEngine, or Replicate
skl-higgsfieldHiggsfield DoP cinematic image-to-video with NSFW handling and async status polling

---

☁️ cloud-providers - Cloud infrastructure management across 8 major providers

SkillWhat it does
skl-cloudflare-dnsDNS zones, records, SSL/TLS, wrangler CLI, bulk operations
skl-cloudflare-workersWorkers, Pages, KV/D1/R2/Durable Objects, Cron triggers
skl-cloudflare-tunnelsTunnels, Zero Trust access policies, WARP, Docker integration
skl-aws-iamIAM users/roles/policies, STS, Secrets Manager, 5 policy templates
skl-aws-computeEC2, Lambda, ECS, App Runner, CDK patterns
skl-aws-storageS3, RDS, DynamoDB, EFS, Backup with bucket policy templates
skl-aws-networkingVPC (3-tier template), Route 53, CloudFront, ALB, VPN, peering
skl-hetznerVPS, dedicated, Object Storage, hcloud CLI, Docker cloud-init
skl-digitaloceanDroplets, App Platform, managed DBs, DOKS, Spaces

---

💬 community - Community management across Discord, Telegram, and Slack

SkillWhat it does
skl-discordDiscord server setup, bots, moderation, growth playbook, analytics, GitHub/CI webhooks
skl-telegramTelegram bots, channels, groups, moderation, notifications, growth, Mini Apps
skl-slackSlack workspace management, app/bot dev (Bolt SDK), Block Kit UI, Workflow Builder, webhooks
**🎬 ai-video-tools** — AI video generation, avatar creation, animated GIFs, explainer production, and multi-platform ad specs
SkillWhat it does
ai-video-generation40+ models via inference.sh CLI: text-to-video, image-to-video, Veo, Wan, Seedance
ai-avatar-lipsyncOmniHuman/Fabric/PixVerse models; audio-driven lipsync and dubbing workflows
remotion-videoReact-based video: compositions, audio, captions, 3D, transitions, cloud render
animated-gif-creatorComposable animation primitives, Slack constraints (2MB/64KB), easing and optimization
explainer-videoScript formulas (AIDA), pacing rules, scene planning, voiceover and music integration
storyboard-creationShot composition, camera angles, movement, continuity; pre-production workflow
video-ad-specsTikTok/Instagram/YouTube/Facebook/LinkedIn dimensions, timing requirements, AIDA framework
pipeline-validationMulti-agent QA gates: specs, narrative, render readiness, asset handoff checks

---

🧊 3d-graphics — 3D performance optimization, asset pipelines, animation principles, and generative art

SkillWhat it does
3d-performanceLOD strategies, frustum/occlusion culling, draw call reduction, R3F-specific optimizations
asset-optimizationgltf-transform pipeline: Draco mesh compression, WebP/KTX2 textures, LOD generation
animation-principlesDisney's 12 animation principles applied to 3D/game contexts, timing and frame count guidelines
algorithmic-artp5.js generative art: seeded randomness, flow fields, particle systems, interactive parameters

---

📱 content-creation — Social strategy, video scripting, storyboarding, and platform ad specs

SkillWhat it does
social-media-marketingYouTube/TikTok/Instagram/LinkedIn growth strategy, thumbnails, influencer workflows
storyboard-creationShot composition, angles, movement, continuity for film and animation pre-production
explainer-videoAIDA script formulas, pacing rules, scene planning, voiceover and music guidance
video-ad-specsPlatform-specific dimensions, timing, and AIDA framework for paid social

---

🤖 ai-ml-dev — AI/ML development, cloud GPU training, and mathematical animation

SkillWhat it does
ai-ml-developmentModel selection, training loops, evaluation, fine-tuning, RLHF, RAG, MLOps patterns
cloud-gpu-trainingRunPod/Lambda/Vast.ai: SCP workflow, batch sizing, cost estimates, checkpointing
manim-animation3Blue1Brown-style math animations: scenes, LaTeX, graphs, algorithm demos in Python

---

🚀 startup-tools — VC fundraising, business formation, product development, and resource access

SkillWhat it does
startup-vc-fundraisingPre-seed through Series C+: pitch decks, investor targeting, term sheets, due diligence
| startup-business-formation | Delaware C-Corp, cap table, 83(b) election, founder vesting, foreign q

Quick Start

Slim (just the core gald3r system — no optional packs)

git clone https://github.com/wrm3/gald3r_template_slim.git 


**Step 2 — Run the installer:**

> **Windows users:** Do not double-click the `.ps1` file — Windows will ask "Open with what?". Instead, open PowerShell and run:
powershell

Configuration

Optional Skill Packs

Beyond the 195 built-in gald3r system skills, skill_packs/ contains 142 optional, domain-specific packs across 20 categories you install on demand. These are not loaded by default — nothing installs until you run the pack's install.ps1. Each pack deploys to all IDE targets (.cursor/, .claude/, .agent/, .codex/, .opencode/).

CategoryPacksDescription
vercel25AI SDK, AI Gateway, Next.js, Next Cache, Turbopack, Routing Middleware, Runtime Cache, Vercel Functions, Blob/Storage, Deployments/CI-CD, Chat SDK, Sandbox, Marketplace, Auth, next-forge, next-upgrade, shadcn, Vercel CLI, Vercel Agent, Vercel Workflow, env-vars, knowledge-update, react-best-practices, bootstrap, verification
cursor-team-kit19compiler-errors-check, cli-control, ui-control, deslop, ci-fix, merge-conflicts-fix, pr-comments-get, ci-loop, pr-easy-review, branch-pr-new, pr-review-canvas, review-ship, smoke-tests-run, verify-this, review-weekly, work-summary, workflow-from-chats, receiving-code-review, requesting-code-review
superpowers13brainstorming, agents-parallel-dispatch, plans-execute, branch-finish, code-review-receive, code-review-request, subagent-development, debug-systematic, tdd, git-worktrees, superpowers-guide, verify-completion, writing-plans, writing-skills
huggingface10hf-cli, hf-dataset-viewer, hf-datasets, hf-evaluation, hf-jobs, hf-model-trainer, hf-paper-publisher, hf-tool-builder, hf-trackio, huggingface-gradio
user-skills12babysit, canvas, hook-create, rule-create, skill-create, subagent-create, cursor-sdk, skills-migrate, shell, prs-split, statusline, cursor-settings-update
ai-video-tools8(AI video generation and editing tools)
cloud-providers16Cloudflare DNS, AWS, GCP, Azure, Hetzner, Oracle Cloud, and more
phantom-connect7social-login-add, phantom-wallet-mcp, sol-transaction-send, browser-app-setup, react-app-setup, react-native-app-setup, message-sign
startup-tools4(Startup toolchain and growth skills)
ai-ml-dev3(AI/ML development and training skills)
ai-media2Seedance 2.0, Higgsfield cinematic video generation
3d-graphics5(3D modeling and rendering skills)
community4(Community building and management skills)
content-creation4(Content writing and marketing skills)
infrastructure4MCP builder, project scaffolding, GitHub integration
blockchain1(Blockchain/Web3 development)
create-plugin2plugin-scaffold-create, review-plugin-submission
context71context7-mcp — live library docs via Context7 MCP
firecrawl1Full web crawling and scraping with LLM-optimized output
continual-learning1Continual learning patterns for agent sessions

```powershell

Command Reference

Task & Bug Management

CommandWhat it does
@g-task-newCreate a task with full spec and acceptance criteria
@g-task-addAdd a task quickly with minimal spec
@g-task-updateUpdate task status, priority, or metadata
@g-task-updUpdate specific task fields
@g-task-delDelete a task and remove from TASKS.md
@g-task-sync-checkValidate TASKS.md ↔ tasks/ file sync
@g-bug-reportLog a bug with severity, file, and description
@g-bug-fixFix a reported bug and update BUGS.md
@g-bug-addAdd a bug entry to BUGS.md
@g-bug-updUpdate an existing bug entry
@g-bug-delDelete a bug entry
@g-goFull autonomous cycle (implement + verify in sequence)
@g-go-codeImplement-only: marks completed items [🔍], never [✅]
@g-go-reviewVerify-only: run in a new agent session to confirm [🔍] items
@g-statusProject status: active tasks, goals, open bugs, ideas
@g-workflowTask expansion and sprint planning

Planning & Goals

CommandWhat it does
@g-planCreate or update PLAN.md and milestone docs
@g-setupInitialize gald3r in a new project
@g-goal-updateUpdate PROJECT_GOALS.md
@g-phase-addAdd a new project phase
@g-phase-pivotPivot project direction
@g-subsystemsSync check, add, update subsystem Activity Logs
@g-subsystem-addAdd a new subsystem to the registry
@g-subsystem-updUpdate subsystem spec or Activity Log
@g-subsystem-delDelete a subsystem from the registry
@g-subsystem-graphGenerate visual subsystem dependency graph
@g-prd-addCreate a new Product Requirements Document (governance/audit artifact)
@g-prd-updUpdate a PRD (blocked on released/superseded status)
@g-prd-delArchive a PRD (soft-delete; audit trail preserved)
@g-prd-reviseCreate v2 of a released PRD; atomically supersedes the original

Feature Pipeline

CommandWhat it does
@g-feat-newStage a new feature and create the feature spec file
@g-feat-addAdd to an existing feature (collect approach, deliverable)
@g-feat-updUpdate feature status, metadata, or notes
@g-feat-promotePromote a specced feature to committed — creates tasks
@g-feat-renameRename a feature slug and update all references
@g-feat-delDelete a feature and remove from FEATURES.md
@g-recon-repo5-pass deep analysis of an external repo → recon report
@g-recon-urlOne-time URL capture into vault
@g-recon-docsCrawl platform docs with freshness tracking
@g-recon-ytExtract and save YouTube transcript
@g-recon-fileAnalyze local files/folders for patterns
@g-res-reviewReview recon report — approve/reject findings
@g-res-deepDeep dive on a specific approved finding
@g-res-applyApply approved findings into feature staging

Knowledge & Vault

CommandWhat it does
@g-vault-ingestManual vault file ingest with frontmatter
@g-vault-searchSearch the vault
@g-vault-statusVault health and freshness report
@g-vault-lintAudit vault frontmatter compliance
@g-vault-process-inboxProcess pending vault inbox items
@g-learnCapture insights to vault memory files

Multi-Project (PCAC)

CommandWhat it does
@g-pcac-adoptRegister a child project (bidirectional link)
@g-pcac-claimRegister a parent project (bidirectional link)
@g-pcac-statusView topology position and open INBOX items
@g-pcac-orderBroadcast a task to child projects with cascade depth
@g-pcac-askSend a request to the parent project
@g-pcac-syncInitiate contract sync with a sibling project
@g-pcac-readReview and action all INBOX items
@g-pcac-notifySend a lightweight FYI notification across topology
@g-pcac-moveTransfer files/folders to another project in topology
@g-pcac-spawnSpawn a new gald3r project — creates, installs gald3r, seeds with features/code, links topology
@g-pcac-send-toSend files, features, ideas, or code to any related project with INBOX notification

Code Quality & Git

CommandWhat it does
@g-code-reviewStructured review: security, performance, quality, architecture
@g-testCreate or run test plans (L1 fast, L2 comprehensive, L3 regression)
@g-crrClean-room rewrite pipeline: analyze external patterns then produce original implementation
@g-git-commitConventional commit with task reference and agent footer
@g-git-sanityPre-commit check: staged secrets, large files, sync drift
@g-git-pushPre-push gate: regular vs release validation

Constraints & Compliance

CommandWhat it does
@g-constraint-addAdd a new architectural constraint
@g-constraint-checkRun compliance check against all active constraints
@g-constraint-updUpdate an existing constraint
@g-constraint-delDelete a constraint

Ideas & Discovery

CommandWhat it does
@g-idea-captureCapture an idea to IDEA_BOARD.md
@g-idea-reviewReview and evaluate IDEA_BOARD entries
@g-idea-farmProactive codebase scan for improvement opportunities

Autonomous Execution

CommandWhat it does
@g-goRun autonomously through the backlog (self-review mode — both phases)
@g-go-codeImplementation-only run — marks tasks [🔍], never [✅]
@g-go-reviewVerification-only run — run in a **new agent session** from the coder
@g-go-swarmCoordinate multi-agent swarm execution across the backlog
@g-go-code-swarmMulti-agent implementation swarm
@g-go-review-swarmMulti-agent review swarm
@g-go-goAutopilot loop - repeatedly calls @g-go until the backlog is clear or all remaining tasks are blocked
@g-missionAutonomous completion loop - runs @g-go iterations until your stated verifiable condition is met; evaluator self-checks each iteration; turn-budgeted
@g-juggernautAlias for @g-mission — unstoppable forward momentum flavor
@g-kamikazeAlias for @g-mission — all-in commitment flavor

Release Management

CommandWhat it does
@g-release-newCreate a new release entry in .gald3r/releases/
@g-release-assignAssign tasks/features to a release
@g-release-statusShow current release status and progress
@g-release-accelerateIdentify what can be fast-tracked to hit a release target
@g-release-publishFinalize and publish a release (CHANGELOG, VERSION, tag)
@g-shipShip a versioned release from [Unreleased] CHANGELOG entries - bumps VERSION, tags, optionally publishes GitHub release

Personality & Skill Pack Management

CommandWhat it does
@g-pers-listList all available personality packs and show which is currently active
@g-pers-pickSwap the active personality pack - removes old rules/skills from all platform dirs, installs new ones
@g-skill-pack-listList available skill packs and per-skill install status across active platform dirs
@g-skill-pack-addInstall a skill pack or individual skill to all active platform dirs
@g-skill-pack-delRemove a skill pack or individual skill from all active platform dirs
@g-skill-pack-saveSave a user-evolved skill back into the pack with _evolved suffix to survive updates

Maintenance

CommandWhat it does
@g-medkit.gald3r/ health check, repair, and version migration
@g-medicTargeted repair for a specific .gald3r/ file or subsystem
@g-tier-setupSet up or upgrade gald3r tier in a project (slim → full → adv)
@g-swot-reviewSWOT analysis for the current project phase
@g-dependency-graphGenerate DEPENDENCY_GRAPH.md from task dependencies
@g-cli-cursorCursor CLI reference and usage patterns
@g-cli-claudeClaude Code CLI reference and usage patterns
@g-cli-geminiGemini CLI reference and usage patterns
@g-cli-codexCodex CLI reference and usage patterns
@g-cli-copilotGitHub Copilot CLI reference and usage patterns

---

🎯 aiskill88 AI 点评 B 级 2026-05-23

该框架提供了丰富的功能和API，适合于AI开发和研究人员使用。但是，框架的文档和示例代码需要进一步完善。

📚 实用指南（长尾问题）
适合谁
  • 使用 Cursor 编辑器、希望提升 AI 编程效率的开发者
  • 需要让 Claude / Cursor 操作本地工具的 AI 工程师
  • 构建多智能体协作系统的 Agent 开发者
  • 构建企业知识库 / RAG 检索应用的团队
  • 跨境业务、多语言内容运营团队
最佳实践
  • 配置 MCP 服务器时建议使用 stdio 传输 + JSON-RPC，避免暴露公网
  • 生产部署优先使用 Docker Compose 隔离依赖，并挂载 volume 持久化数据
  • 分块大小建议 256-512 tokens，向量库优选 pgvector 或 Qdrant
  • Agent 任务先做 dry-run 验证工具调用链，再开启自主执行
  • Cursor rules 控制在 80 行内，否则模型上下文成本会显著上升
常见错误
  • API key 直接提交到 git 仓库（请用 .env 并加入 .gitignore）
  • MCP 配置路径拼错或权限不足，重启 Claude Desktop 才生效
  • 容器内无法访问宿主机 localhost — 使用 host.docker.internal
  • embedding 模型与查询模型不一致导致检索失效
  • Python 依赖冲突：建议用 venv / uv 隔离环境
部署方案
  • Docker：gald3r 提供官方镜像，docker compose up 一键启动
  • CLI：直接 npm install -g / pip install，命令行调用
  • 云端托管：可放在 Vercel / Railway / Fly.io 等 PaaS 平台
相关搜索
gald3r 中文教程gald3r 安装报错怎么办gald3r MCP 配置gald3r Docker 部署gald3r Agent 工作流gald3r 与同类工具对比gald3r 最佳实践gald3r 适合谁用
⚡ 核心功能
👥 适合谁
  • 使用 Cursor 编辑器、希望提升 AI 编程效率的开发者
  • 需要让 Claude / Cursor 操作本地工具的 AI 工程师
  • 构建多智能体协作系统的 Agent 开发者
  • 构建企业知识库 / RAG 检索应用的团队
⭐ 最佳实践
  • 配置 MCP 服务器时建议使用 stdio 传输 + JSON-RPC，避免暴露公网
  • 生产部署优先使用 Docker Compose 隔离依赖，并挂载 volume 持久化数据
  • 分块大小建议 256-512 tokens，向量库优选 pgvector 或 Qdrant
  • Agent 任务先做 dry-run 验证工具调用链，再开启自主执行
⚠️ 常见错误
  • API key 直接提交到 git 仓库（请用 .env 并加入 .gitignore）
  • MCP 配置路径拼错或权限不足，重启 Claude Desktop 才生效
  • 容器内无法访问宿主机 localhost — 使用 host.docker.internal
  • embedding 模型与查询模型不一致导致检索失效
👥 适合人群
自动化工程师和运维人员项目经理和业务分析师希望减少重复性工作的专业人士数字化转型团队
🎯 使用场景
  • 自动化日常重复性工作，将精力集中于创造性任务
  • 构建数据采集 → 处理 → 输出的完整自动化管线
  • 实现跨平台、跨系统的数据流转和业务协同
⚖️ 优点与不足
✅ 优点
  • +大幅减少重复性人工操作
  • +可视化流程，清晰直观
  • +可扩展性强，支持复杂场景
⚠️ 不足
  • −初始配置和调试需投入一定时间
  • −强依赖外部服务的稳定性
  • −复杂场景需具备一定技术基础
⚠️ 使用须知

该工具使用 NOASSERTION 协议，商用场景请仔细阅读协议条款，必要时咨询法律意见。

AI Skill Hub 为第三方内容聚合平台，本页面信息基于公开数据整理，不对工具功能和质量作任何法律背书。

建议在沙箱或测试环境中充分验证后，再部署至生产环境，并做好必要的安全评估。

📄 License 说明

📄 NOASSERTION — 请查阅原始协议条款了解具体使用限制。

🔗 相关工具推荐
LLM资源合集（精选）
精选100+可直接运行的AI Agent和RAG应用集合。包含完整工作流示例、智能代理框架和检索增强生成系统。适合AI开
LangChain AI开发框架
Agent工作流
ai-agents-for-beginners Agent工作流
微软官方开源项目，提供12堂系统课程学习AI智能体框架。涵盖工作流设计、RAG检索增强、多智能体协作等核心技能。适合AI
n8n AI工作流自动化
Agent工作流
📰 相关 AI 新闻
AI Agent 自主化能力最新进展
AI 资讯 · 知识关联
AI Agent 自主化能力最新进展
AI 资讯 · 知识关联
AI Agent 自主化能力最新进展
AI 资讯 · 知识关联
GitHub热门：claude-mem — Persistent Context Across Sessions for Every Agent – Capt
AI 资讯 · 知识关联
🍿 AI 圈相关吃瓜
AutoGPT 自主完成了任务：把我的文件夹全部重命名了
AI 圈观察
配了5个 MCP 工具，Claude 一个都没用
AI 圈观察
Filesystem MCP 帮 Claude 找文件，找了整个 node_modules
AI 圈观察
🗺️ 相关解决方案
mcp
mcp-server-setup
agent
ai-agent-workflow
workflow
ai-workflow-templates
translation
ai-translation-pipeline
🧩 你可能还需要
基于当前 Skill 的能力图谱，自动补全的工具组合
开源AI工具：RAG知识库系统
基于Vue.js前端的RAG知识库系统，提供高效的知识检索和生成功能，助力AI应用开发
KohakuTerrarium Agent工作流
通用型AI代理框架，提供开箱即用的工作流编排能力。支持多种LLM集成，适合开发者构建复杂智能应用、自动化任务流程和多步骤AI任务编排。
CrewAI 多代理协作平台
MCP · Agent · 工作流
技能寻求者
MCP · Agent · 工作流
total-agent-memory MCP工具
为Claude Code和Codex CLI提供持久化记忆功能的开源MCP工具。自动提取知识图谱，支持多轮对话上下文保留，适合需要长期记忆和
natively-cluely-ai-assistant — Claude Skill 中文使用文档
免费开源的AI面试助手，实时转录，隐蔽模式，局部RAG，BYOK。无订阅，防止数据泄露。
❓ 常见问题 FAQ
使用pip安装：pip install gald3r
💡 AI Skill Hub 点评

经综合评估，开源AI工作流 在Agent工作流赛道中表现稳健，质量良好。如果你已有明确的使用需求，可以直接上手体验；如果还在评估阶段，建议对比同类工具后再做决策。

⬇️ 获取与下载
📚 深入学习 开源AI工作流
查看分步骤安装教程和完整使用指南，快速上手这款工具
⚙️ 安装教程 📚 使用教程
🌐 原始信息
原始名称 gald3r
原始描述 开源AI工作流：AI dev framework: task management, 23 agents, 116+ skills, 50+ commands. Runs on。⭐17 · Python
Topics workflowagent-frameworkaiai-agentsai-memoryaiderpython
GitHub https://github.com/wrm3/gald3r
License NOASSERTION
语言 Python
🔗 原始来源
🐙 GitHub 仓库  https://github.com/wrm3/gald3r 🌐 官方网站  https://www.gald3r.ai/

收录时间：2026-05-19 · 更新时间：2026-05-24 · License：NOASSERTION · AI Skill Hub 不对第三方内容的准确性作法律背书。