Compression
thedaviddias/Front-End-Checklist
A skill your agent uses when auditing slow page loads, heavy assets, or rendering delays related to Enable text-based compression.
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。
Apache-2.0Auto-check passed
Install LLM Call Sites
$ npx skills add stello-agent/stello --skill llm-call-sites -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install stello-agent/stello llm-call-sites --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/llm-call-sites .claude/skills/llm-call-sites && 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 "llm-call-sites" agent skill from https://github.com/stello-agent/stello/tree/main/.agents/skills/llm-call-sites into .claude/skills/llm-call-sites/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llm-call-sites", 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/llm-call-sitesType 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 llm-call-sites -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install stello-agent/stello llm-call-sites --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/llm-call-sites .agents/skills/llm-call-sites && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "llm-call-sites" agent skill from https://github.com/stello-agent/stello/tree/main/.agents/skills/llm-call-sites into .agents/skills/llm-call-sites/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llm-call-sites", 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 llm-call-sites -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install stello-agent/stello llm-call-sites --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/llm-call-sites .cursor/skills/llm-call-sites && 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 "llm-call-sites" agent skill from https://github.com/stello-agent/stello/tree/main/.agents/skills/llm-call-sites into .cursor/skills/llm-call-sites/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llm-call-sites", 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/llm-call-sites--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 llm-call-sites -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install stello-agent/stello llm-call-sites --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/llm-call-sites .gemini/skills/llm-call-sites && 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 "llm-call-sites" agent skill from https://github.com/stello-agent/stello/tree/main/.agents/skills/llm-call-sites into .gemini/skills/llm-call-sites/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llm-call-sites", 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 llm-call-sitesInstalls 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 llm-call-sites -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/llm-call-sites .github/skills/llm-call-sites && 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 "llm-call-sites" agent skill from https://github.com/stello-agent/stello/tree/main/.agents/skills/llm-call-sites into .github/skills/llm-call-sites/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llm-call-sites", 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 llm-call-sites -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 llm-call-sites --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/llm-call-sites .opencode/skills/llm-call-sites && 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 "llm-call-sites" agent skill from https://github.com/stello-agent/stello/tree/main/.agents/skills/llm-call-sites into .opencode/skills/llm-call-sites/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llm-call-sites", 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
llm-call-sites- GitHub stars
- 112
- Token cost
- ~793 tokens
- SKILL.md length
- 186 words
- Files
- 1
- Skills in repo
- 12
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
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。
- Works in 4 steps: Session 对话 → Compress → Consolidate（L3 → memory） → …
- SKILL.md covers 1. Session 对话, 2. Compress, 3. Consolidate（L3 → memory） and 4. Reflection（应用层自行实现）, plus 2 more sections
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
LLM Call Sites is an agent skill from 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。
Its SKILL.md is about 790 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
- “/llm-call-sites”
Workflow steps
4 steps, taken from the step headings in SKILL.md.
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.
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
LLM Call Sites loads about 793 tokens when it runs. Until then it costs about 29 tokens; SKILL.md has 186 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). 186 words, ~793 tokens.
Download SKILL.mdSave it as
.claude/skills/llm-call-sites/SKILL.md (or your agent's skills folder).- name
- llm-call-sites
- description
- 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。
LLM 调用消息结构
Stello 里所有 LLM 调用的 messages 参数构成。Session 同构——root 与 child 走同一套上下文组装规则，差异只在 meta.label。
1. Session 对话
[
{ role: 'system', content: systemPrompt }, // 可能含 <parent_context> 块；若非空
{ role: 'system', content: <session_identity> }, // 若 meta.label 非空
{ role: 'system', content: insight }, // 若非空，消费后清除
{ role: 'system', content: compressSummary }, // 仅当触发自动压缩
...recentL3History, // user / assistant / tool
{ role: 'user', content: userInput },
]tools 经 llm.stream(messages, { tools }) 第二参数传入，不进 messages。Session.stream() 直接转发文本增量，Session.send() 消费同一条 stream 后返回聚合结果。
<session_identity> 形态（label 缺省则该消息不注入）：
<session_identity>
你当前在「{meta.label}」子会话中。
</session_identity>label 改名后下次 send 自动同步，无需重写持久化的 systemPrompt。
systemPrompt 在 fork-compress 场景形态：
{合成后的 systemPrompt}
<parent_context>
{父 session 压缩摘要}
</parent_context>所有 Session 同构走这套规则。Root 也是普通 Session，差异只在 meta.label。如需在 root 上注入"全局综合"，应用层把综合结果通过 putInsight(rootId, content) 一次性写入即可。memory 槽位不进入 send() 上下文。
2. Compress
[
{ role: 'system', content: COMPRESS_PROMPT },
{ role: 'system', content: <role_context> }, // 若传入非空 roleContext
{ role: 'user', content: "对话记录:\n" + messages.map(m => `${m.role}: ${m.content}`).join('\n') },
]两种触发：
- 对话内自动压缩（超阈值 80%）：
messages= 待压缩的 L3 头部 - fork 时父→子（
context: 'compress'）：messages= 父 session 全量 L3
输出：纯文本摘要。
3. Consolidate（L3 → memory）
[
{ role: 'system', content: CONSOLIDATE_PROMPT },
{ role: 'system', content: <role_context> }, // 若传入非空 roleContext
{ role: 'user', content: [
currentMemory ? `当前摘要:\n${currentMemory}` : null,
`对话记录:\n${messages.map(m => `${m.role}: ${m.content}`).join('\n')}`,
].filter(Boolean).join('\n\n') },
]currentMemory= 本 session 当前 memory 槽位messages= 本 session 全量 L3
输出：100-150 字摘要，写回本 session memory 槽位。
4. Reflection（应用层自行实现）
跨 Session 的"全 memory → 综合 → 定向 insight"循环由应用层自行调用任意 LLM 完成：
- 输入：
agent.listSessionDigests({ status: 'active' })——{ id, label, memory, insight }[] - 输出：派生 per-target
insight，通过agent.putInsight(targetId, content)写回
应用层完全掌控 prompt 形态、调用频率、LLM tier。详见 stello-agent-creation §7 与 stello-agent-usage §6.4。
XML Tag 注入汇总
| Tag | 调用路径 | 数据来源 | 注入位置 |
|---|---|---|---|
<parent_context> | Session 对话（仅 fork-compress 场景） | 父 session 压缩摘要 | 合成进 systemPrompt 字段 |
<session_identity> | Session 对话 | SessionMeta.label | systemPrompt 之后 |
<role_context> | Compress / Consolidate | DefaultFnOptions.roleContext（应用层传入） | 任务 prompt 之后、user content 之前 |
共性
| 维度 | 对话类（1） | 提炼类（2、3） |
|---|---|---|
| 接口 | llm.stream(msgs, { tools }) | 内部消费 llm.stream(msgs) 后聚合为 string |
| tools | 有 | 无 |
| L3 形态 | 原始 message 数组 | ${role}: ${content} 字符串拼接进 user content |
| 返回 | 结构化（含 tool calls） | 纯文本 |
<think> 清洗 | 否 | 是 |
© 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/llm-call-sites of stello-agent/stello.
Open the folder on GitHubat commit 3bc9493
Compare with similar skills
LLM Call Sites 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 |
|---|---|---|---|---|---|---|
| LLM Call Sites this skillstello-agent/stello | 112 | — | ~793 | Automated safety check: Pass | Apache-2.0 | |
| Compressionthedaviddias/Front-End-Checklist | 74k | — | ~421 | Automated safety check: Pass | MIT | |
| Sitesasgeirtj/system_prompts_leaks | 69k | — | ~1.6k | Automated safety check: Pass | CC0-1.0 | |
| Image Compressionthedaviddias/Front-End-Checklist | 74k | — | ~405 | Automated safety check: Pass | MIT | |
| OmniRoute Prompt Compressiondiegosouzapw/OmniRoute | 75k | — | ~1.6k | Automated safety check: Pass | MIT | |
| Site Mdgoogle-labs-code/stitch-skills | 8.5k | — | ~783 | Automated safety check: Pass | Apache-2.0 |
Similar skills
Sites
asgeirtj/system_prompts_leaks
A skill your agent uses when creating or updating a website, web app, or browser game, or when a visual layout or interactive tool would help with what the user is doing.
Image Compression
thedaviddias/Front-End-Checklist
A skill your agent uses when reviewing image assets, markup, and CDN or build transforms related to Compress images without quality loss.
OmniRoute Prompt Compression
diegosouzapw/OmniRoute
Configures OmniRoute's RTK, Caveman and stacked compression modes, manages language packs and rules, and previews token savings of 60 to 90 percent.
- Official
Site Md
google-labs-code/stitch-skills
Analyze a project description or site requirements and synthesize a project constitution into SITE.md for the Stitch Build Loop
Static Sites
bionic-gpt/bionic-gpt
Create or modify Bionic's generated marketing site, documentation, blog, course pages, and static assets.
More from stello-agent/stello
All 12 skills in this repoFork Design
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…
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 LLM Call Sites
What does LLM Call Sites do?
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。. LLM Call Sites is an agent skill from stello-agent/stello.
How do I install LLM Call Sites in Claude Code?
Run `npx skills add stello-agent/stello --skill llm-call-sites -a claude-code`. Or copy the skill folder (.agents/skills/llm-call-sites in stello-agent/stello) into .claude/skills/llm-call-sites in your project. Claude Code loads it when a task matches its description.
How do I install LLM Call Sites in Codex?
Run `npx skills add stello-agent/stello --skill llm-call-sites -a codex`. Or copy the skill folder (.agents/skills/llm-call-sites in stello-agent/stello) into .agents/skills/llm-call-sites in your project. Codex loads it when a task matches its description.
Can I use LLM Call Sites 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 llm-call-sites -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/llm-call-sites, .gemini/skills/llm-call-sites, .github/skills/llm-call-sites and .opencode/skills/llm-call-sites in your project.
What does LLM Call Sites need to run?
SKILL.md names no scripts, command-line tools or credentials: LLM Call Sites is instructions for the agent only.
Does LLM Call Sites 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 LLM Call Sites 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 LLM Call Sites use?
LLM Call Sites 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 LLM Call Sites use?
About 793 tokens (SKILL.md is roughly 3.2k 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 LLM Call Sites?
Skills that share tags, products or a category with LLM Call Sites: Compression (thedaviddias/Front-End-Checklist, 74k stars), Sites (asgeirtj/system_prompts_leaks, 69k stars), Image Compression (thedaviddias/Front-End-Checklist, 74k stars) and OmniRoute Prompt Compression (diegosouzapw/OmniRoute, 75k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains LLM Call Sites?
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.