跳转到主要内容
此内容尚未提供您的语言版本,正在以英文显示。

Metal

技能 已验证 活跃

Extract the conceptual essence of a repository as skills, agents, and teams — the project's roles, procedures, and coordination patterns expressed as agentskills.io-standard definitions. Reads an arbitrary codebase and produces generalized definitions that capture WHAT the project does and WHO operates it, without replicating HOW it does it. Use when onboarding to a new codebase and wanting to understand its conceptual architecture, when bootstrapping an agentic system from an existing project, when studying a project's organizational DNA for cross-pollination, or when creating a skill/agent/team library inspired by a reference implementation.

目的

To transform implicit project structure and logic into explicit, reusable definitions for agentic systems, onboarding, and knowledge transfer.

功能

  • Extracts conceptual essence of repositories
  • Produces agentskills.io-standard definitions for skills, agents, and teams
  • Identifies roles, procedures, and coordination patterns
  • Focuses on project's 'what' and 'who', not 'how'
  • Supports various output depths and scope limitations

使用场景

  • Onboarding to new codebases to understand conceptual architecture
  • Bootstrapping agentic systems from existing projects
  • Studying organizational DNA for cross-pollination
  • Creating skill/agent/team libraries inspired by reference implementations

非目标

  • Replicating the project's codebase or implementation details
  • Generating executable code or detailed step-by-step instructions for specific tasks
  • Analyzing code quality or security vulnerabilities (focus is conceptual essence)

工作流

  1. Prospect: Survey repository structure, read self-description, identify project boundary.
  2. Assay: Sample representative files, identify conceptual domains, verbs, roles, and flows, classify as essential/accidental.
  3. Meditate: Release implementation bias, re-evaluate findings against the Ore Test, abstract to higher levels.
  4. Smelt: Classify essential concepts into generalized skills, agents, or teams with descriptive names and one-line descriptions.
  5. Heal: Verify extraction quality, check for over/under-extraction, generalization, and balance.
  6. Cast: Produce agentskills.io-standard definitions for skills, agents, and teams.
  7. Temper: Final validation of count, coverage, independence, and pass/fail the Ore Test on all items.

先决条件

  • Path to the repository or project root directory
  • Purpose statement for essence extraction
  • Optional: Focus domains
  • Optional: Output depth and maximum extractions

安装

/plugin install agent-almanac@pjt222-agent-almanac

质量评分

已验证
100 /100
about 23 hours ago 分析

信任信号

最近提交2 days ago
星标14
许可证MIT
状态
查看源代码

类似扩展

Lean Ctx

100

AI 代理的上下文运行时 — 包含 59 个 MCP 工具、10 种读取模式、95+ 种 shell 模式、支持 18 种语言的 tree-sitter AST。将 LLM 上下文压缩高达 99%。用于读取文件、运行 shell 命令、搜索代码或探索目录。如果不存在,则自动安装。

技能
yvgude

Pathfinder

100

将代码库映射为按功能分组的流程图,识别不同功能之间的重复关注点,并提出统一的架构。在被要求“寻找理想路径”、统一重复系统或在重构前审计架构时使用。输出一个建议的统一流程图以及针对每个系统的“制定计划”提示。

技能
thedotmack

Codacy Audit

100

Codacy Cloud workflow for this repository -- run Codacy's analyzers locally before `git push` (mirrors what Codacy CI runs), and fetch/cluster Codacy issues for any PR via the v3 API. Use when the user mentions Codacy, "codacy analysis", `codacy-analysis-cli`, "codacy issues on PR", "fix codacy CI", "codacy markdownlint findings", or any Codacy gate failing on a netdata-org PR. Ships scripts analyze-local.sh (docker/binary runner for codacy-analysis-cli) and pr-issues.sh (paginated v3 issue fetch + group-by tool/pattern/severity/file). Token-safe -- CODACY_TOKEN never reaches assistant-visible stdout. Read-only by design in the current SOW; write actions (mark FP, mark fixed) are deferred.

技能
netdata

Domain Extract

100

Extract domain knowledge from existing project sources and generate domain rules. Also handles vault sync and domain listing.

技能
luiseiman

Auto Optimize

100

自动化分析、评估、设计和优化目标项目。将 VibeGuard 集成作为基线扫描,修复过程遵循 VibeGuard 规范,并在最后运行合规性检查。支持 auto-run-agent 自主执行。

技能
majiayu000

Understand Domain

100

从代码库中提取业务领域知识并生成交互式领域流程图。可独立运行(轻量级扫描)或从现有的 /understand 知识图中派生。

技能
Lum1104