What makes Skill Alchemy Main notable?

1.1K downloads. Trust evidence available

How should Skill Alchemy Main be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

Skill Alchemy Main

SkillAlchemy — 一念落地，万象成形。输入任意想法或蒸馏目标，输出可安装的 SKILL.md。内部编排 Lens（看清问题）和 LEAP（执行蒸馏/融合）。用户唯一入口。 Use when 用户说「蒸馏」「生成 skill」「融合」「我想做 X 但不知道从哪下手」。 Skill: Skill Alchemy Main Owner: agentsope Summary: SkillAlchemy — 一念落地，万象成形。输入任意想法或蒸馏目标，输出可安装的 SKILL.md。内部编排 Lens（看清问题）和 LEAP（执行蒸馏/融合）。用户唯一入口。 Use when 用户说「蒸馏」「生成 skill」「融合」「我想做 X 但不知道从哪下手」。 Tags: latest:0.1.3 Version history: v0.1.3 | 2026-06-15T11:41:28.437Z | user Fixed model names in benchmark table. v0.1.2 | 2026-06-12T14:15:37.330Z | user Updated README with SkillsBench benchmark results. v0.1.1 | 2026-06-02T

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.1k

Updated

Oct 11, 2026

Version

0.1.3

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.1K downloads reported by the source. Last updated 10/11/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 11, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 11, 2026
Adoption signal: 1.1K downloadsadoption · observed Oct 11, 2026
Latest release: 0.1.3release · observed Jun 15, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s172fz0df5jxmx2a90ds9q6zfd877b2a:skillalchemy

Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.
Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-agentsope-skillalchemy/snapshot"

Documentation

CLAWHUB

157,277 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: SkillAlchemy
description: |
  SkillAlchemy — 一念落地，万象成形。输入任意想法或蒸馏目标，输出可安装的 SKILL.md。
  内部编排 Lens（看清问题）和 LEAP（执行蒸馏/融合）。用户唯一入口。
  Use when 用户说「蒸馏」「生成 skill」「融合」「我想做 X 但不知道从哪下手」。
version: v1.0
---

# Skill-Alchemy · 一念落地，万象成形

你是 SkillAlchemy。编排两个子 skill：Lens 看清，LEAP 落地。
你自己不蒸馏、不融合——只做编排。**所有用户交互由你负责，LEAP 不跟用户说话。**

## 前置检查

```
ls ~/.claude/skills/Lens/SKILL.md
ls ~/.claude/skills/LEAP/SKILL.md
```

**如果缺少任何一个，告诉用户：**

> SkillAlchemy 需要两个依赖才能运行，请先安装：
>
> ```
> npx skills add agentsope/SkillAlchemy/skills/Lens
> npx skills add agentsope/SkillAlchemy/skills/LEAP
> ```
>
> 或者去 https://skills.sh 搜索 Lens 和 LEAP 安装。
>
> 装好之后回来找我继续。

---

## 编排流程

### Phase 0: 确认深度 + 任务简报

先确认 depth。用户没说就问一句：

```
quick    — 快速原型，3 agent，~5-8 min，跳过验证
standard — 日常使用（默认），4-5 agent，~15-20 min
deep     — 发布级，6-8 agent，~25-35 min，强制验证 + 双审核
没说的话默认 standard。
```

用户给了深度后，**展示任务简报：**

```
◆ 任务简报

▸ 需求    蒸馏「张雪峰」→ persona skill
▸ 流程    Lens → A 分支（7 Stage + 2 Gate）
          ├─ Research Swarm  4-5 agent 并行研究
          ├─ Exemplar        find-skills 在线检索 + 自动评分
          └─ Compile         编译 + 自评 + 验证 + 清理
▸ 深度    standard · ~15-20 min
▸ 交互    步步确认（2 次暂停）

> 确认，按 standard 跑
> 换成 deep，研究更深入、验证更严格、双 agent 交叉审核
> 一路默认跑完，中间别问我了，全部默认值到底
> 先只要 Lens 看看维度，不生成 skill
```

根据实际任务替换内容。确认后进 Phase 1。如果用户一开始就指定了 depth，跳过询问直接出简报。

**「一路默认」模式：** 用户在任何节点说「一路默认」→ 跳过当前及后续所有交互，全部 standard 默认值跑完。

---

### Phase 1: Lens 分析

调 Lens，输入用户原话。Lens 不向用户提问，直接输出增强版 description。

**Lens 完成后，展示维度摘要（不放全文，太长）：**

```
◆ Lens 分析完成 · N 个维度

  [维度名]    [维度名]    [维度名]
  [维度名]    [维度名]    [维度名]
  ...

▸ 意图    distill_persona / distill_method / fuse_skills

> 确认，进入 [distill / fuse] 管线继续
> 展开看看完整的 Lens 分析原文，每个维度的细节
> 补一个 XX 维度，重新分析一遍
> 就停在这，我消化一下 Lens 的结果，不继续了
```

确认后进 Phase 2。提了修改意见 → 重新调 Lens 带上反馈。
「一路默认」已激活 → 跳过，直接进 Phase 2。

---

### Phase 2: 路由判断

| Lens 意图 | 动作 |
|-----------|------|
| distill | → Phase 3a（A 分支：蒸馏管线） |
| fuse | → Phase 3b（B 分支：融合管线） |
| decompose | 停。展示 Lens 输出，问是否继续 |
| 无法判断 | 问用户：蒸馏还是融合？ |

---

### Phase 3: 执行

**所有输出落在当前项目根目录的 `output/` 下。**
调 LEAP 时用绝对路径指定输出位置（以实际项目路径为准）。

#### 3a. Distill 路线（2 步，1 次确认）

**Step 1: 生成 research plan。**
```
调 LEAP：
  "distill [target]，depth [depth]。
   只到 research plan（stop_after_stage: 3），
   输出到 <项目根目录>/output/<target>-skill/。"
```

LEAP 跑完 Stage 1-3 后停止。读取 `research_plan.json`：

```
◆ Research Plan · N agents

  R1  [维度名]
      [搜索方向一句话]

  R2  [维度名]
      [搜索方向一句话]

  ...

> 确认，按这个计划启动 N 个 agent 并行研究
> 加一个 R[n] 专门研究 XX 方向，补上缺失的维度
> 删掉 R[n]，这个维度我不太关心，省点资源
> 换成 quick 快速跑，3 个 agent 够了我赶时间
```

**Step 2: 研究 + exemplar + 编译（无交互，直接跑完）。**
```
调 LEAP：
  "从 Stage 4 继续 distill [target]，
   research_plan 已确认，
   输出到 <项目根目录>/output/<target>-skill/。"
```

LEAP 执行 Stage 4-7 + Gate 1-2，全自动完成：
Research Swarm → Exemplar Discovery（find-skills + score_skill 自动评分择优）→ Synthesis → Compile → Validate。

完成后清理中间产物：
- 删除 `references/exemplar_candidates.json`（临时评分文件）
- 删除 `references/exemplars/`（中间参照副本）
- 删除空 `validation/`（standard 模式不跑 Phase 8）
- 保留 `R*.md`（研究证据）、`intermed

skills/agentsop-agent-topology-selection/SKILL.md

---
name: agentsop-agent-topology-selection
version: 0.1.0
description: >-
  Cross-framework enhancement overlay for choosing a multi-agent topology BEFORE writing any
  agent. A binary-question rubric — is single-agent + tools enough? do agents need to know
  about each other? does the output need one voice? — maps the answer to single-agent /
  supervisor / swarm / sequential / hierarchical. Activates when a coder agent is tempted to
  "split the work into roles" or reaches for a multi-agent framework. Encodes the *selection
  rubric* that the per-framework skills assume but never surface. Search keywords: when to
  use multi-agent, single vs multi agent, do I need multiple agents, supervisor vs swarm,
  multi-agent vs single agent, agent team design.
overlay: true
cross_links: [crewai, langgraph, bounded-loop]
---

# Multi-Agent Topology Selection · SOP (ENHANCE overlay)

> Overlay posture: this skill decides *whether and which* topology. It does not
> teach the API — descend to `[[crewai]]` or `[[agentsop-langgraph]]` for that. Every
> load-bearing claim carries an inline source tag resolving in
> `references/R1-source-evidence.md`.

---

## 1. 何时激活 (When to Activate)

Activate when **any** of the following fire:

- The task description contains "team of agents", "researcher + writer + reviewer",
  "manager agent", "agents that hand off", "split this into roles", or "multi-agent".
- A coder agent is about to instantiate ≥2 agents (CrewAI `Agent(...)` × N,
  LangGraph supervisor/swarm, OpenAI Swarm handoffs) and has **not yet** justified
  why a single agent with tools is insufficient.
- Someone is choosing between CrewAI `Process.sequential` vs `Process.hierarchical`,
  or LangGraph supervisor vs swarm vs hierarchical-teams, and wants the *rubric*,
  not the syntax.
- A multi-agent system is over budget on tokens/latency and the question is "can we
  collapse agents back into one?".

Do **not** activate for: a single LLM call, a one-shot RAG query, or a fixed
tool-call pipeline with no role separation. Those are the single-agent baseline
this skill defends.

> Mental check: *"An agent needs agency, otherwise it's just another script."*
> — João Moura, CrewAI founder `[[crewai · §1.3]]`. If you can write the control
> flow in `if/else`, you do not need multiple agents — you need one agent (or a
> graph) with explicit edges.

---

## 2. 核心心智模型 (Core Mental Model)

**Most "multi-agent" problems are single-agent + tools.** Add agents only when
*context isolation* or *parallel expertise* genuinely demands it.

> "Single-agent is right for approximately 80% of cases; the trap is reaching for
> multi-agent because it sounds more capable." `[[crewai · DC-1]]`

Two — and only two — forces justify a second agent:

1. **Context isolation.** One agent's working context would pollute another's
   (a critic that must not see its own draft's rationalisations; a tool-heavy
   sub-task whose 40 intermediate tool calls should not bloat the main thread).
   Spl

skills/agentsop-aider/SKILL.md

---
name: agentsop-aider
version: 1.0.0
description: >-
  SOP for terminal-based, git-native AI pair programming with Aider (git work-tree + tree-sitter repo-map + edit-format + human-in-loop REPL). Use when editing code in an existing git repo via an LLM, when you need to converge a change to 2-5 files, pick an edit format that fits the model, run architect+editor mode, or wire an auto-test loop.
domain: terminal-based AI pair programming, git-native code editing
source: aider.chat docs + Paul Gauthier's blog + leaderboards
audience: coder-agents and human engineers who edit code via LLMs
---

# Aider SOP — 终端结对编程的操作系统

> 一句话：Aider 是“**git 工作树 + tree-sitter 仓库地图 + 编辑格式 + 人类在环 REPL**”的四元组。理解这四个原语，剩下的都是配置。

## 1. 何时激活本技能

下列任一情形成立时，按本 SOP 进入 Aider 工作模式：

- 任务是**编辑已有 git 仓库**里的代码（不是从零起项目）。
- 你能把改动范围**收敛到 2–5 个文件**，或愿意先用 `/ask` 让模型借助 repo-map 把范围找出来。
- 你需要**逐步可回滚**的修改历史（每次编辑一个 commit，`/undo` 一步回退）。
- 你在**终端**里工作（tmux / 远程 ssh / CI）；或者你在写一个把 Aider 当子进程驱动的 agent。
- 你关心**编辑格式对模型质量的影响**（diff / udiff / whole / patch 的选择问题）。
- 你需要 BYOM（自带模型），跑本地 LLM 或非主流厂商。

**不应激活的反面信号**：见 §6 反模式与边界。

## 2. 核心心智模型

### 2.1 四个原语

```
+------------------+   +------------------+   +------------------+   +------------------+
| 1. Git working   |   | 2. Tree-sitter   |   | 3. Edit format   |   | 4. REPL loop     |
|    tree          |   |    repo-map      |   |    (wire proto)  |   |    (你在环里)    |
|                  |   |                  |   |                  |   |                  |
| - per-edit       |   | - symbol-level   |   | - diff / udiff   |   | - /ask /code     |
|   commit         |   |   summary        |   |   / whole /      |   |   /architect     |
| - /undo          |   | - PageRank over  |   |   patch          |   | - 每轮人手确认   |
| - dirty 文件     |   |   import graph   |   | - 模型适配选择   |   | - 不自主         |
|   先 commit 再编 |   | - 动态预算       |   | - JSON 是反模式  |   |                  |
+------------------+   +------------------+   +------------------+   +------------------+
```

四者缺一不可：
- 去掉 git → 失去回滚与审计；
- 去掉 repo-map → 大仓库里 LLM 找不到正确文件（SWE-Bench Lite 上 repo-map 让 Aider 70.3% 命中正确文件 [aider.chat/2024/05/22/swe-bench-lite.html]）；
- 用错 edit format → 出现“lazy coding”、SEARCH 块找不到、JSON 句法破坏（udiff 在 GPT-4 Turbo 上把 refactor 基准从 20% 拉到 61% [aider.chat/2023/12/21/unified-diffs.html]）；
- 放弃 REPL → 退化为自主 agent，但 Aider 在 SWE-Bench 上恰好证明“人在环 + 多次尝试”比纯自主链路更稳。

### 2.2 LLM 看到的上下文分三层（优先级递减）

| 层 | 内容 | 谁能改 |
|---|---|---|
| 系统提示 + 编辑格式说明 | Aider 固化 | Aider |
| 只读上下文 | repo-map + `/read` 文件 + CONVENTIONS.md | 你（通过 `--read`） |
| 读写上下文 | `/add` 的文件 | LLM **只能编辑**这里的文件 |

> **铁律**：LLM 只允许编辑 `/add`-ed 的文件。这是 Aider 的安全边界。模型“改错了文件”几乎总是因为该文件没 `/add` 或你 `/add` 了太多无关文件。

### 2.3 上下文预算（25k 信号阈）

> "Above about 25k tokens of context, most models start to become distracted." [aider.chat/docs/troubleshooting/edit-errors.html]

把这条当硬约束：超过 25k tokens，编辑准确率断崖式下降。`/tokens` 持续监控。

### 2.4 Repo-map 不是 RAG

repo-map 是 **tree-sitter 提取的符号清单**（类、函数、签名），用 PageRank 在源文件依赖图上排序，**塞给 LLM 当地图**。这不是 embedd

skills/agentsop-bio-fraud-forensics/SKILL.md

---
name: agentsop-bio-fraud-forensics
domain: research-integrity
trigger_keywords:
  - "data fraud / image manipulation"
  - "Western blot duplication / splicing"
  - "GRIM / statcheck / impossible statistics"
  - "paper mill / tortured phrases"
  - "PubPeer / Retraction Watch verification"
description: >-
  Screens biomedical / life-science papers for signs of data fabrication, image
  manipulation, and statistical anomalies, using the detection techniques distilled
  from the field's canonical exposure platforms (PubPeer, Data Colada, Science
  Integrity Digest, For Better Science) and tools (ImageTwin/Proofig, statcheck,
  GRIM/GRIMMER, Problematic Paper Screener, Seek & Blastn). Use when asked to check
  a paper/figure for image duplication, blot splicing, impossible statistics, paper-mill
  or tortured-phrase signals, research integrity, or "is this data faked"; or when a
  user shares a figure, Western blot, supplementary dataset, or DOI and asks whether it
  looks manipulated. Reports observable anomalies as questions for clarification — it
  never accuses anyone of fraud.
version: 1.0.0
---

# Bio-Fraud Forensics · 生物医学论文数据造假筛查

A screening methodology for life-science papers. It reverse-engineers how real cases
were caught — the exact panels compared, the transform applied, the statistic recomputed —
and turns that into a reproducible per-paper checklist. It is a **detective's lens, not a
verdict machine**: every output stays at "observed anomaly" or "question for the authors,"
because red flag ≠ proof and an accusation can end a career.

## Activation Rules

**Trigger when:**
- "Check this paper / figure / Western blot for manipulation," "does this data look faked," "screen for image duplication."
- A user shares a figure, blot, microscopy panel, supplementary `.xlsx`, or a DOI and asks if it's trustworthy.
- "Is this a paper mill?", "tortured phrases," "are these statistics possible," "run GRIM/statcheck on this."
- "Where do I check if this paper has been flagged / retracted?" (verification routing).
- Asked to draft a PubPeer-grade, reproducible image/data integrity comment.

**Do NOT trigger when:**
- The user wants a scientific peer review of validity/novelty (use a peer-review skill) rather than an integrity screen.
- The user asks you to publicly accuse a named person of fraud, or to write an accusation/social post (refuse — see Boundary Rules).
- The task is general statistics help or figure-making with no integrity question.
- The paper is non-biomedical and the request is about a domain whose fraud signatures differ (physics/CS); say so and scope down.

## Agentic Protocol

Run this as a chain-of-steps. Cheapest, fastest signals first; the expensive image/stat
forensics last (they tell you *where* to dig is often answered for free by the cheap checks).

**Step 1 — Scope & status.** Identify the input: single figure, full paper, supplementary
dataset, or a batch. Run the status cascade in parallel (it's free and may hand you t

skills/agentsop-bounded-loop/SKILL.md

---
name: agentsop-bounded-loop
version: 0.1.0
description: >-
  Universal discipline for any LM-driven loop — agent retries, plan-act-observe, multi-agent
  handoffs, optimiser passes, test-fix cycles. Encodes the one rule every framework
  documents quietly and every team relearns expensively: the LM in the loop is NEVER a
  reliable terminator. Termination must be provided by an explicit counter + exit predicate
  + stagnation signal + escalation path that live OUTSIDE the LM's control. This is a tool-
  level, framework-agnostic skill. It maps onto LangGraph (recursion_limit + state counter +
  interrupt), CrewAI (max_iter + max_rpm + human_input), Claude / OpenAI SDKs
  (max_iterations + tool_use_budget), DSPy (declared evaluation budget), Aider (REPL +
  explicit retry cap), and AutoGen (max_consecutive_auto_reply). Search keywords: infinite
  loop, recursion limit, recursion_limit, GraphRecursionError, max iterations, max_iter,
  agent stuck, agent won't stop, runaway agent, ReAct loop not terminating, agent repeating
  itself.
---

# bounded-loop · O7

> Source posture: every load-bearing claim is cited inline with a short tag
> resolved against `references/R1-source-evidence.md` and
> `references/R2-cross-framework.md`. Examples cite the real GitHub issues
> they're distilled from.

---

## 1. 何时激活 (Activation Rules)

Activate this skill when **any** of the following is true:

- The task involves a workflow that contains a **cycle** — tool-call → reflect
  → retry, plan → act → observe → re-plan, draft → critique → revise,
  test → fix → re-test.
- The user is hitting a framework's "loop too deep" error:
  `GRAPH_RECURSION_LIMIT` (LangGraph), `MaxIterationsExceeded` (LangChain
  `AgentExecutor`), "agent exceeded max_iter" (CrewAI), `max_turns reached`
  (OpenAI Agents SDK), `stop_reason="max_tokens"` mid-tool-use (Anthropic).
- The user proposes "let's just raise the limit" / "set max_iter to 100" /
  `recursion_limit=200` — this is the canonical anti-pattern this skill
  exists to prevent.
- The user is building a **multi-agent** system with delegation, handoff,
  or supervisor patterns — these are exposure-multipliers for unbounded
  loops (see `[gh/crewai-330]`).
- The user is building an **optimiser / evaluator loop** (DSPy, AutoEval,
  RLHF, self-refining agent) where "stop when good enough" is the
  termination criterion — this is *never* sufficient on its own.
- The user wants a **test-fix loop**, **self-healing code agent**, or
  **iterative refinement** workflow — every code-agent in production
  (Cursor, Aider, Devin, Claude Code) ships with an explicit step budget.

Do **not** activate for: single LLM calls, one-shot RAG queries, stateless
tool pipelines, or flows where the cycle is provably bounded by data (e.g.,
"iterate once per row in this fixed list").

---

## 2. 核心心智模型 (Core Mental Model)

**Every loop body must produce a state change that proves progress — and
the proof must be checkable without calling another LM.**

Github ReposUpdated 2d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.