Fork
alpha-omega-security/scrutineer
Stage a scanned repository into a private repo in the configured GitHub organisation.
Fork 机制完整说明。覆盖 ForkProfile 与 EngineForkOptions 的字段对齐、四层 fallback 合成链（sessionDefaults → parent → profile → forkOptions）、systemPrompt 合成三种模式、skills 三态语义、持久化边界（SerializableSessionConfig 只固化…
Apache-2.0Auto-check passed
Install Fork Design
$ npx skills add stello-agent/stello --skill fork-design -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install stello-agent/stello fork-design --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/stello-agent/stello.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/fork-design .claude/skills/fork-design && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "fork-design" agent skill from https://github.com/stello-agent/stello/tree/main/.agents/skills/fork-design into .claude/skills/fork-design/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "fork-design", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/stello-agent/stello/tree/main/.agents/skills/fork-designType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add stello-agent/stello --skill fork-design -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install stello-agent/stello fork-design --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/stello-agent/stello.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/fork-design .agents/skills/fork-design && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "fork-design" agent skill from https://github.com/stello-agent/stello/tree/main/.agents/skills/fork-design into .agents/skills/fork-design/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "fork-design", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add stello-agent/stello --skill fork-design -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install stello-agent/stello fork-design --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/stello-agent/stello.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/fork-design .cursor/skills/fork-design && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "fork-design" agent skill from https://github.com/stello-agent/stello/tree/main/.agents/skills/fork-design into .cursor/skills/fork-design/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "fork-design", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/stello-agent/stello.git --path .agents/skills/fork-design--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add stello-agent/stello --skill fork-design -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install stello-agent/stello fork-design --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/stello-agent/stello.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/fork-design .gemini/skills/fork-design && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "fork-design" agent skill from https://github.com/stello-agent/stello/tree/main/.agents/skills/fork-design into .gemini/skills/fork-design/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "fork-design", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install stello-agent/stello fork-designInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add stello-agent/stello --skill fork-design -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/stello-agent/stello.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/fork-design .github/skills/fork-design && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "fork-design" agent skill from https://github.com/stello-agent/stello/tree/main/.agents/skills/fork-design into .github/skills/fork-design/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "fork-design", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add stello-agent/stello --skill fork-design -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install stello-agent/stello fork-design --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/stello-agent/stello.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/fork-design .opencode/skills/fork-design && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "fork-design" agent skill from https://github.com/stello-agent/stello/tree/main/.agents/skills/fork-design into .opencode/skills/fork-design/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "fork-design", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
fork-design- GitHub stars
- 112
- Token cost
- ~2k tokens
- SKILL.md length
- 629 words
- Files
- 1
- Skills in repo
- 12
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Fork 机制完整说明。覆盖 ForkProfile 与 EngineForkOptions 的字段对齐、四层 fallback 合成链（sessionDefaults → parent → profile → forkOptions）、systemPrompt 合成三种模式、skills 三态语义、持久化边界（SerializableSessionConfig 只固化…
- Works in 8 steps: 校验 profile 存在性 → SplitGuard 拦截（可选） → 读取 sourceSession 的固化配置 → …
- SKILL.md covers 定位, 触发路径, 类型体系 and 四层 Fallback 合成链, plus 5 more sections
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
Fork Design is an agent skill from stello-agent/stello. Fork 机制完整说明。覆盖 ForkProfile 与 EngineForkOptions 的字段对齐、四层 fallback 合成链（sessionDefaults → parent → profile → forkOptions）、systemPrompt 合成三种模式、skills 三态语义、持久化边界（SerializableSessionConfig 只固化 systemPrompt/skills）。任何涉及 fork / profile / stellocreatesession / 配置合成 的工作都应读这个。
Its SKILL.md is about 2k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
The repository describes itself as: Conversations aren't linear — why should AI chats be? The first open-source conversation topology engine. Auto-branching session trees, inherited memory, star-map visualization… The licence is Apache-2.0.
Example prompts
- “/fork-design”
Workflow steps
8 steps, taken from the first numbered list in SKILL.md.
- 校验 profile 存在性
- SplitGuard 拦截（可选）
- 读取 sourceSession 的固化配置
- 按四层链合成最终 SessionConfig
- 创建拓扑节点（topology-first，先拿 ID）
- 将可序列化子集固化入存储
- 调用 session.fork({ id, ... }) 创建 Session 实例
- 触发 onSessionFork 事件
What it can do on your machine
Read from SKILL.md and the folder at commit 3bc9493. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md (its code samples are typescript).
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Fork Design loads about 2k tokens when it runs. Until then it costs about 70 tokens; SKILL.md has 629 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from stello-agent/stello at commit 3bc9493, republished under its Apache-2.0 licence (© stello-agent). 629 words, ~1,968 tokens.
Download SKILL.mdSave it as
.claude/skills/fork-design/SKILL.md (or your agent's skills folder).- name
- fork-design
- description
- Fork 机制完整说明。覆盖 ForkProfile 与 EngineForkOptions 的字段对齐、四层 fallback 合成链（sessionDefaults → parent → profile → forkOptions）、systemPrompt 合成三种模式、skills 三态语义、持久化边界（SerializableSessionConfig 只固化 systemPrompt/skills）。任何涉及 fork / profile / stello_create_session / 配置合成 的工作都应读这个。
Fork — 子 Session 创建机制
定位
Fork 是 Stello 创建子 Session 的唯一路径。编排层（Engine）负责完整编排：
- 校验 profile 存在性
- SplitGuard 拦截（可选）
- 读取 sourceSession 的固化配置
- 按四层链合成最终 SessionConfig
- 创建拓扑节点（topology-first，先拿 ID）
- 将可序列化子集固化入存储
- 调用
session.fork({ id, ... })创建 Session 实例 - 触发
onSessionFork事件
调用方不感知内部顺序。两条触发路径（LLM / 代码）共用这整套流程。
触发路径
| 触发者 | 入口 | 典型场景 |
|---|---|---|
| LLM | 内置 tool stello_create_session | LLM 判断当前任务需要拆分为子任务 |
| 应用层代码 | agent.forkSession(sourceId, options) | 代码驱动的主动编排（UI 按钮、策略脚本） |
两条路径汇聚到同一个 forkSession(options: EngineForkOptions)。LLM 路径只是在前面加了一层 JSON Schema 校验 + tool args → EngineForkOptions 映射。
类型体系
三个类型层层叠加，共享同一个基座。
基座：SessionConfig（6 字段）
所有 Session 运行时可配置项。由 sessionDefaults、父 session 固化 config、ForkProfile、EngineForkOptions 共同描述，mergeSessionConfig 输出也是这个形状。
| 字段 | 类型 | 说明 |
|---|---|---|
systemPrompt | string | Session 的 system prompt |
llm | LLMAdapter | LLM 适配器 |
tools | LLMCompleteOptions['tools'] | 用户 tool 定义 |
skills | string[] | skill 白名单（三态语义见下文） |
consolidateFn | SessionCompatibleConsolidateFn | L3→L2 提炼函数 |
compressFn | SessionCompatibleCompressFn | 上下文压缩函数 |
ForkProfile extends SessionConfig
预注册的 fork 配置模板，注册期定义。在 6 个基座字段之上新增 4 个 fork 专属字段：
| 新增字段 | 类型 | 说明 |
|---|---|---|
systemPromptFn | (vars) => string | 动态模板，优先于 systemPrompt 字段 |
systemPromptMode | 'preset' | 'prepend' | 'append' | 合成策略，默认 'prepend' |
context | 'none' | 'inherit' | ForkContextFn | 上下文继承策略（默认值） |
prompt | string | fork 后的开场消息（默认值） |
EngineForkOptions extends SessionConfig
运行时每次 fork 传入的参数。在 6 个基座字段之上新增 6 个运行时字段：
| 新增字段 | 类型 | 说明 |
|---|---|---|
label | string（必填） | 子 session 显示名 |
prompt | string | fork 后的开场消息 |
context | 'none' | 'inherit' | ForkContextFn | 上下文继承（覆盖 profile 默认值） |
topologyParentId | string | 显式指定拓扑父节点（不传 = 当前 sessionId） |
profile | string | 引用预注册的 ForkProfile 名 |
profileVars | Record<string, string> | systemPromptFn 的模板变量 |
字段对齐矩阵
| 字段 | SessionConfig | ForkProfile | EngineForkOptions |
|---|---|---|---|
| systemPrompt | ✓ | ✓（静态） | ✓ |
| llm / tools / skills | ✓ | ✓ | ✓ |
| consolidateFn / compressFn | ✓ | ✓ | ✓ |
| systemPromptFn | — | ✓ | — |
| systemPromptMode | — | ✓ | — |
| context | — | ✓（默认） | ✓（覆盖） |
| prompt | — | ✓（默认） | ✓（覆盖） |
| label | — | — | ✓ 必填 |
| topologyParentId | — | — | ✓ |
| profile / profileVars | — | — | ✓ |
Profile 和 Options 的差异都是职责驱动的：profile 是模板（不能自引用 profile，不需要 label），options 是运行时参数（需要 label 每次给出，需要引用 profile 入口）。
四层 Fallback 合成链
所有 fork 最终都要合成一份完整的 SessionConfig 交给 session.fork()。合成输入有四层：
sessionDefaults → parent（固化 config） → profile → forkOptions
低优先级 高优先级字段级覆盖规则：后层非 undefined 的字段覆盖前层；undefined 永不覆盖（保留前层值）。
普通字段（llm / tools / skills / consolidateFn / compressFn）
直接走 later-wins 链。每层独立决定某字段是否贡献。
systemPrompt 特殊合成
分有无 profile 两种情况：
情况 A — 有 profile：
- 先求 profile 的 promptSource：
profile.systemPromptFn?.(profileVars) ?? profile.systemPrompt - 按
profile.systemPromptMode（默认'prepend'）合成 promptSource 与forkOptions.systemPrompt：
| Mode | 结果 |
|---|---|
'preset' | 仅用 profilePrompt；forkOptions 的 systemPrompt 被忽略 |
'prepend'（默认） | {profilePrompt}\n\n{forkOptionsPrompt} |
'append' | {forkOptionsPrompt}\n\n{profilePrompt} |
- 若 profile + forkOptions 都未贡献 prompt（如 preset 模式两者皆空），回落到 parent → defaults 的普通 later-wins 链。
情况 B — 无 profile：
走 [defaults, parent, forkOptions] 的普通 later-wins 链。
skills 三态语义
skills 不做合并，整数组替换。三种取值：
| 取值 | 含义 |
|---|---|
undefined | 未配置，本层不贡献 — 继承下层值；若所有层皆 undefined，运行时继承全局 SkillRouter（无白名单） |
[] | 显式禁用 — 该 session 的 activate_skill 看不到任何 skill，可覆盖下层非空值 |
['a', 'b'] | 白名单 — 只允许这几个 skill 可见 |
显式 [] 覆盖下层 ['a','b'] 是标准行为（"undefined 不覆盖"不阻止显式空数组生效）。
实战场景
| 场景 | 合成链贡献 |
|---|---|
| 从 root session fork | parent 层 = root 的 SerializableSessionConfig（root 是普通 session，正常参与合成链） |
| 从非 root session fork | parent 层 = 该 session 的 SerializableSessionConfig（只有 systemPrompt/skills） |
| 无 profile 的普通 fork | profile 层 = undefined |
| Profile + options 都提供 llm | 结果取 options 的 llm |
| Profile 提供 llm，options 不提供 | 结果取 profile 的 llm |
Show full SKILL.md (245 more words)Show less
持久化边界
SerializableSessionConfig 只固化两个字段：systemPrompt、skills。
原因：其余四个字段（llm / tools / consolidateFn / compressFn）本质是运行时引用（函数、适配器、闭包），不可安全序列化。
持久化时机
forkSession() 完成合成后：
- 从合成结果挑
systemPrompt和skills两字段打包为SerializableSessionConfig - 若两字段都是
undefined（空对象），跳过putConfig写入，避免给存储层制造噪声 - 否则写入
sessions.putConfig(childId, serializable)
运行时重建时的后果
Engine 重新装配某 session 的 runtime config 时：
systemPrompt/skills从固化存储重放llm/tools/consolidateFn/compressFn不来自父 session 的持久化，而是每次 fork 时从sessionDefaults → profile → forkOptions现场合成
实际语义：嵌套 fork（子再 fork 孙）时，孙 session 的 llm 不会自动继承子 session 的 llm。孙 session 的 llm 来自 sessionDefaults（或孙 fork 时显式指定的 profile/options）。要让某条分支始终用特殊 llm，需在每次 fork 时显式传入、或通过 profile 固化。
Fork 专属行为
topologyParentId vs sourceSessionId
两者分离是编排层的关键设计：
| 概念 | 含义 | 来源 |
|---|---|---|
sourceSessionId | 上下文来源 session（系统提示词、历史继承的对象） | 总是 = 当前 session.id |
topologyParentId | 拓扑父节点（星空图上挂靠位置） | options 显式给出，默认 = sourceSessionId |
不传 topologyParentId 时两者相等（默认树形拓扑：fork from X → 挂在 X 下）。调用方显式传入 topologyParentId 可让两者分离——例如想把节点挂到根或任意已有节点下，但上下文继承仍来自发起 fork 的 session（sourceSessionId = current）。
context 继承策略
| 值 | 含义 |
|---|---|
'none'（默认） | 子 session 以空对话历史启动 |
'inherit' | 拷贝父 session 全部 L3 记录 |
ForkContextFn | 自定义函数，接收父消息数组返回继承子集 |
options.context 优先级高于 profile.context。
prompt 开场消息
Fork 后写入子 session 的首条 assistant 消息，用户进入子 session 时首先看到。options.prompt 优先于 profile.prompt。
LLM 侧暴露 — stello_create_session
Engine 构造时自动注入此内置 tool。参数 schema 随 ForkProfileRegistry 状态动态变化：
基础参数（始终存在）：label（必填）、systemPrompt、prompt、context（enum: 'none' | 'inherit'）
条件参数（仅当有 profile 注册时追加）：
profile：enum，取值为所有已注册 profile 名vars：object，键值对字符串，传给systemPromptFn
LLM 调用后，Engine 把 tool args 映射为 EngineForkOptions，走和代码路径完全相同的 forkSession() 流程。
Profile 注册模式
固定角色（静态 prompt）
profiles.register('poet', {
systemPrompt: '你是一位诗人，所有回复用诗歌形式。',
systemPromptMode: 'preset',
})preset 模式让 LLM 传入的 systemPrompt 被忽略，保证角色不被覆盖。
动态模板（基于变量）
profiles.register('region-expert', {
systemPromptFn: (vars) => `你是${vars.region}地区的留学专家。`,
systemPromptMode: 'preset',
skills: ['search', 'summarize'],
})调用方传 profileVars: { region: '北美' } 生成具体 prompt。
基础约束（允许追加）
profiles.register('researcher', {
systemPrompt: '你是研究助手，善于深入分析。',
systemPromptMode: 'prepend', // 允许 fork options 追加具体研究主题
context: 'inherit',
})LLM 在 prepend 模式下可通过 systemPrompt 参数补充具体任务约束，合成结果为 {profile}\n\n{task-specific}。
设计不变量（不会改的决策）
- 三类型同根 — SessionConfig / ForkProfile / EngineForkOptions 共享 6 字段基座；职责驱动的差异字段独立声明
- 四层顺序固定 — defaults → parent → profile → forkOptions，不允许调换或插入新层
- undefined 不覆盖 — 保证"某层不传"等价于"使用下层值"的直觉
- skills 显式
[]能生效 — 与 undefined 区分，让"禁用"成为可表达的意图 - 只固化 systemPrompt + skills — 可序列化字段有限，其余字段每次 fork 现场合成
- root 是普通 session — root 的固化 systemPrompt/skills 通过 parent 层正常进入子 session 的合成链，没有任何特殊豁免
- topologyParentId 与 sourceSessionId 分离 — 编排层的拓扑策略和上下文继承是两个独立维度
© stello-agent, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in .agents/skills/fork-design of stello-agent/stello.
Open the folder on GitHubat commit 3bc9493
Compare with similar skills
Fork Design next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Fork Design this skillstello-agent/stello | 112 | — | ~2k | Automated safety check: Pass | Apache-2.0 | |
| Forkalpha-omega-security/scrutineer | 231 | — | ~3.3k | Automated safety check: Pass | MIT | |
| Forking Pathsbrycewang-stanford/Auto-Empirical-Research-Skills | 4.5k | — | ~2.4k | Automated safety check: Pass | Custom licence | |
| Refresh Forkmarin-community/marin | 3.9k | — | ~4.1k | Automated safety check: Pass | Apache-2.0 | |
| Session Source Forkmvschwarz/openrig | 5.9k | — | ~2.2k | Automated safety check: Pass | Apache-2.0 | |
| Release Latex Forkzly2006/zhihu-plus-plus | 4.2k | — | ~1.8k | Automated safety check: Pass | AGPL-3.0 |
Similar skills
Forking Paths
brycewang-stanford/Auto-Empirical-Research-Skills
Map the garden of forking paths for a concrete econometric design and turn it into a machine-readable design card that the specification-search engine can walk.
Refresh Fork
marin-community/marin
Refresh a named Marin external fork pin onto a newer upstream base using its configured descriptor and required end-to-end test.
Session Source Fork
mvschwarz/openrig
A skill your agent uses when authoring a rig spec member or rig expand payload that needs to start a new managed seat from a prior runtime conversation source — sessionsource: { mode: fork, ref: {…
Release Latex Fork
zly2006/zhihu-plus-plus
Release the LaTeX fork used by Zhihu++. An agent skill from zly2006/zhihu-plus-plus.
Fork Fleet
aeonfun/aeon
Fork divergence monitor - tracks where the fleet's active forks diverge in CODE (unique commits, new/modified skills) and CONFIG (enable/var/model/schedule vs upstream), gated on real change.
More from stello-agent/stello
All 12 skills in this repoLLM Call Sites
stello-agent/stello
Stello 框架内所有 LLM 调用位置的消息结构速查。覆盖 Session 对话、compress、consolidate；应用层 reflection 调用由 orchestrator 自行决定。
Session Usage
stello-agent/stello
Session 对话单元的设计理念、上下文组装规则、memory / insight 槽位语义、单一 Session 模型与跨 Session 通信模型。
Stello Agent Creation
stello-agent/stello
StelloAgent 创建配置教程。完整说明 createStelloAgent 的每个配置项，包含 sessionDefaults、storage、tools、skills、forkProfiles、session 层接入、orchestration 等。
Stello Agent Usage
stello-agent/stello
StelloAgent 运行时使用教程。覆盖 Session 生命周期、createSession、turn/stream 对话、fork 配置合成链、orchestrator-facing 数据 SDK、runtime 管理、热更新等运行时 API。
Stello Usage
stello-agent/stello
Stello 仓库总览入口。快速理解各包的关系、推荐入口、编排模型、单一 Session 模型. An agent skill from stello-agent/stello.
Storage Design
stello-agent/stello
存储接口的设计原则、SessionMeta 与 TopologyNode 解耦、上下文槽位、单一 SessionStorage 接口。触发条件：理解或实现 SessionStorage / SessionTree。
Questions about Fork Design
What does Fork Design do?
Fork 机制完整说明。覆盖 ForkProfile 与 EngineForkOptions 的字段对齐、四层 fallback 合成链（sessionDefaults → parent → profile → forkOptions）、systemPrompt 合成三种模式、skills 三态语义、持久化边界（SerializableSessionConfig 只固化…. Fork Design is an agent skill from stello-agent/stello.
How do I install Fork Design in Claude Code?
Run `npx skills add stello-agent/stello --skill fork-design -a claude-code`. Or copy the skill folder (.agents/skills/fork-design in stello-agent/stello) into .claude/skills/fork-design in your project. Claude Code loads it when a task matches its description.
How do I install Fork Design in Codex?
Run `npx skills add stello-agent/stello --skill fork-design -a codex`. Or copy the skill folder (.agents/skills/fork-design in stello-agent/stello) into .agents/skills/fork-design in your project. Codex loads it when a task matches its description.
Can I use Fork Design in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add stello-agent/stello --skill fork-design -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/fork-design, .gemini/skills/fork-design, .github/skills/fork-design and .opencode/skills/fork-design in your project.
What does Fork Design need to run?
SKILL.md names no scripts, command-line tools or credentials: Fork Design is instructions for the agent only.
Does Fork Design access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Fork Design safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Fork Design use?
Fork Design is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Fork Design use?
About 2k tokens (SKILL.md is roughly 7.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Fork Design?
Skills that share tags, products or a category with Fork Design: Fork (alpha-omega-security/scrutineer, 231 stars), Forking Paths (brycewang-stanford/Auto-Empirical-Research-Skills, 4.5k stars), Refresh Fork (marin-community/marin, 3.9k stars) and Session Source Fork (mvschwarz/openrig, 5.9k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Fork Design?
stello-agent (a GitHub organization) maintains it in stello-agent/stello, which has 112 GitHub stars. The repository holds 12 skills in this directory. The repository was last updated on July 24, 2026.
Source: stello-agent/stello on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.