Menu

Not in the top 1,000 by GitHub stars

levi-qiao/longgraph-skill

Long-horizon agent skill for Claude Code / Cursor / Codex / Grok Build — multi-task ledger loop, host-portable, clean-context supervisor, verified gates. Markdown library (loop-graph), not a framework.

Claude CodeCodexEngineering

History

Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.

View weekly changes

SKILL.md

Week of Nov 2, 2026 · Line counts unavailable

Before Not checked yet → After Network or broad shell

First observation → 3e26e19b91e6

No earlier version was observed for comparison.

Compare on GitHub

skills/loop-converge/SKILL.md

Week of Nov 2, 2026 · Line counts unavailable

Before Not checked yet → After Instructions only

First observation → 3e26e19b91e6

No earlier version was observed for comparison.

Compare on GitHub

skills/loop-deliver/SKILL.md

Week of Nov 2, 2026 · Line counts unavailable

Before Not checked yet → After Instructions only

First observation → 3e26e19b91e6

No earlier version was observed for comparison.

Compare on GitHub

skills/loop-graph/SKILL.md

Week of Nov 2, 2026 · Line counts unavailable

Before Not checked yet → After Network or broad shell

First observation → 3e26e19b91e6

No earlier version was observed for comparison.

Compare on GitHub

skills/loop-research/SKILL.md

Week of Nov 2, 2026 · Line counts unavailable

Before Not checked yet → After Instructions only

First observation → 3e26e19b91e6

No earlier version was observed for comparison.

Compare on GitHub

View on GitHubAdd to my skills

Verified commit 3e26e19b91e62dba63c0ab53ad5ba8a670403abb

git clone --filter=blob:none --no-checkout https://github.com/levi-qiao/longgraph-skill.git .hotskills/levi-qiao-longgraph-skill && git -C .hotskills/levi-qiao-longgraph-skill checkout --detach 3e26e19b91e62dba63c0ab53ad5ba8a670403abb # Copy each selected skill directory from .hotskills/levi-qiao-longgraph-skill into .claude/skills/<skill-name>/
git clone --filter=blob:none --no-checkout https://github.com/levi-qiao/longgraph-skill.git .hotskills/levi-qiao-longgraph-skill && git -C .hotskills/levi-qiao-longgraph-skill checkout --detach 3e26e19b91e62dba63c0ab53ad5ba8a670403abb # Copy each selected skill directory from .hotskills/levi-qiao-longgraph-skill into .agents/skills/<skill-name>/
78GitHub stars
Not availablestars in 7 days
5skills in this repository
0commits in 30 days

Heat

Insufficient history

Not enough history yet to measure heat. It appears after a few days of data.

Heat never changes the order by stars, and sponsorship never touches it. How heat works

Stars over time

Nov 4, 2026Nov 5, 202678

30-day change not yet available: not enough history.

Individual skills in this repository

How it works

longgraph

When your agent uses it

Use only when the user explicitly invokes /longgraph or names this skill. Routes and authors a durable longgraph run. Do not use unless named. Do not execute or resume generated runtime node files.

What it loads into context

Frontmatter

54 tokens

Always available to the agent

SKILL.md

402 tokens

When the skill is activated

Reference documents

17,165 tokens

Loaded on request

Token counts are estimates. Scripts run separately; they are not counted as reference context.

The playbook

  1. Fit and route

Scripts executed separately

  • install.sh

Files

  • .claude-plugin/marketplace.jsonasset481 bytes
  • .claude-plugin/plugin.jsonasset899 bytes
  • .github/ISSUE_TEMPLATE/bug_report.mddoc555 bytes
  • .github/ISSUE_TEMPLATE/feature_request.mddoc636 bytes
  • .github/PULL_REQUEST_TEMPLATE.mddoc567 bytes
  • .github/workflows/grok-bot-issue-relay.ymlasset1,202 bytes
  • .github/workflows/grok-bot-pr-relay.ymlasset1,169 bytes
  • .gitignoreasset165 bytes
  • AGENTS.mddoc6,419 bytes
  • CLAUDE.mddoc199 bytes
  • CONTRIBUTING.mddoc4,885 bytes
  • LICENSEasset1,066 bytes
  • README.mddoc15,459 bytes
  • README.zh-CN.mddoc14,835 bytes
  • SKILL.mdcore1,785 bytes
  • SKILL.mdcore1,785 bytes
  • SKILL.mdcore1,785 bytes
  • assets/graph.pngasset626,328 bytes
  • docs/public-private-boundary.mddoc5,903 bytes
  • install.shruns2,876 bytes
  • lib/methodology.mddoc21,651 bytes
  • llms.txtdoc2,849 bytes
Read SKILL.md

# longgraph — route to the right durable run

longgraph has one shared compiler, [`loop-graph`](skills/loop-graph/SKILL.md),
and focused entries that bind only goal-specific rules. All compile the same
executor, supervisor, ledger, directives, and ops artifacts. Do not execute the
generated runtime nodes from this authoring skill.

## Fit and route

- Use [`loop-converge`](skills/loop-converge/SKILL.md) for multi-round unused
  code, duplication, consolidation, reuse, or slimming.
- Use [`loop-deliver`](skills/loop-deliver/SKILL.md) for a product or engineering
  requirement that needs incremental implementation and acceptance proof.
- Use [`loop-research`](skills/loop-research/SKILL.md) to compare feasible
View the complete source on GitHub

Skills included

  • longgraph
  • loop-converge
  • loop-deliver
  • loop-graph
  • loop-research

Repository facts

License
MIT
Created
Jul 19, 2026
Last pushed
Sep 23, 2026
Forks
8
Open issues
0
Project activity
Active

Before you install

Network or broad shell

Automated signals, not a guarantee

Checked Nov 5, 2026

Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.

Show 27 notes
  • SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-converge/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-converge/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-converge/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-deliver/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/loop-graph/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
Show all 84 signals by file

.claude-plugin/marketplace.json

  • Network accessline 4
    "owner": {     "name": "levi-qiao",     "url": "https://github.com/levi-qiao"
  • Network accessline 5
    "name": "levi-qiao",     "url": "https://github.com/levi-qiao"   },
  • Network accessline 6
    "url": "https://github.com/levi-qiao"

.claude-plugin/plugin.json

  • Network accessline 1
    {   "$schema": "https://json.schemastore.org/claude-code-plugin-manifest.json",
  • Network accessline 2
    "$schema": "https://json.schemastore.org/claude-code-plugin-manifest.json",
  • Network accessline 6
    "author": {     "name": "levi-qiao",     "url": "https://github.com/levi-qiao"
  • Network accessline 7
    "name": "levi-qiao",     "url": "https://github.com/levi-qiao"   },
  • Network accessline 8
    "url": "https://github.com/levi-qiao"
  • Network accessline 9
    },   "homepage": "https://github.com/levi-qiao/longgraph-skill",   "repository": "https://github.com/levi-qiao/longgraph-skill",
  • Network accessline 10
    "homepage": "https://github.com/levi-qiao/longgraph-skill",
  • Network accessline 11
    "repository": "https://github.com/levi-qiao/longgraph-skill",

.github/ISSUE_TEMPLATE/bug_report.md

  • Network accessline 24
    **Environment** Claude Code version, OS, and how you installed (curl / git clone / manual).
  • Network accessline 25
    Claude Code version, OS, and how you installed (curl / git clone / manual).

.github/workflows/grok-bot-issue-relay.yml

  • Access to secretsline 19
    [Credential-related source omitted]
  • Network accessline 26
    --arg event_action "$EVENT_ACTION" \             '{source:"github-actions",event_name:$event_name,event_action:$event_action,event:$event}' \           | curl -sS -f -X POST "$WEBHOOK_URL" \
  • Network accessline 27
    [Credential-related source omitted]
  • Network accessline 28
    | curl -sS -f -X POST "$WEBHOOK_URL" \
  • Access to secretsline 29
    [Credential-related source omitted]

.github/workflows/grok-bot-pr-relay.yml

  • Access to secretsline 17
    [Credential-related source omitted]
  • Network accessline 24
    --arg event_action "$EVENT_ACTION" \             '{source:"github-actions",event_name:$event_name,event_action:$event_action,event:$event}' \           | curl -sS -f -X POST "$WEBHOOK_URL" \
  • Network accessline 25
    [Credential-related source omitted]
  • Network accessline 26
    | curl -sS -f -X POST "$WEBHOOK_URL" \
  • Access to secretsline 27
    [Credential-related source omitted]

AGENTS.md

  • Network accessline 94
    - **Conventional commits**: `feat(loop-graph): …`, `fix(loop-graph): …`, `docs: …`. - **Branch + PR**, don't push straight to `main` (this file is the exception the owner asked for). PR body states the failure mode the change addresses or t
  • Network accessline 95
    - **Branch + PR**, don't push straight to `main` (this file is the exception the owner asked for). PR body states the failure mode the change addresses or the tuning it documents. - **Distribution**: installable via the curl script *and* th
  • Network accessline 96
    - **Distribution**: installable via the curl script *and* the Claude Code plugin. The

CONTRIBUTING.md

  • Network accessline 30
    The graph grows one node at a time — a node is **one Markdown prompt + one inspectable edge**, no runtime. The scout ([#17](https://github.com/levi-qiao/longgraph-skill/issues/17) → [#18](https://github.com/levi-qiao/longgraph-skill/pull/18

README.md

  • Network accessline 13
    [![GitHub stars](https://img.shields.io/github/stars/levi-qiao/longgraph-skill?style=flat-square&color=6C63FF)](https://github.com/levi-qiao/longgraph-skill/stargazers)
  • Network accessline 14
    [![License: MIT](https://img.shields.io/badge/License-MIT-14B8A6?style=flat-square)](LICENSE)
  • Network accessline 15
    [![PRs welcome](https://img.shields.io/badge/PRs-welcome-22C55E?style=flat-square)](CONTRIBUTING.md)
  • Network accessline 16
    ![Hosts: Claude Code · Cursor · Codex · Grok Build](https://img.shields.io/badge/Hosts-Claude%20Code%20·%20Cursor%20·%20Codex%20·%20Grok%20Build-111827?style=flat-square)
  • Network accessline 17
    ![Type: agent skill · prompt library](https://img.shields.io/badge/Type-agent%20skill%20·%20prompt%20library-0EA5E9?style=flat-square)
  • Dangerous shell operationline 166
    curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh
  • Network accessline 166
    curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh
  • Network accessline 246
    [public-Git self-iteration case](skills/loop-graph/examples/self-iteration-longgraph-skill/README.md) records how the method was hardened into this library. Special thanks to [@BrightProgrammer7](https://github.com/BrightProgrammer7) for th
  • Network accessline 247
    records how the method was hardened into this library. Special thanks to [@BrightProgrammer7](https://github.com/BrightProgrammer7) for the `migrate-blob-storage` example and the discussions that sharpened milestone
  • Network accessline 248
    [@BrightProgrammer7](https://github.com/BrightProgrammer7) for the
  • Network accessline 254
    [MIT](LICENSE) © 2026 [levi-qiao](https://github.com/levi-qiao)

README.zh-CN.md

  • Network accessline 13
    [![GitHub stars](https://img.shields.io/github/stars/levi-qiao/longgraph-skill?style=flat-square&color=6C63FF)](https://github.com/levi-qiao/longgraph-skill/stargazers)
  • Network accessline 14
    [![License: MIT](https://img.shields.io/badge/License-MIT-14B8A6?style=flat-square)](LICENSE)
  • Network accessline 15
    [![欢迎 PR](https://img.shields.io/badge/PRs-welcome-22C55E?style=flat-square)](CONTRIBUTING.md)
  • Network accessline 16
    ![宿主：Claude Code · Cursor · Codex · Grok Build](https://img.shields.io/badge/Hosts-Claude%20Code%20·%20Cursor%20·%20Codex%20·%20Grok%20Build-111827?style=flat-square)
  • Network accessline 17
    ![类型：agent skill · 提示词库](https://img.shields.io/badge/Type-agent%20skill%20·%20prompt%20library-0EA5E9?style=flat-square)
  • Dangerous shell operationline 153
    curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh
  • Network accessline 153
    curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh
  • Network accessline 226
    [公开 Git 自迭代案例](skills/loop-graph/examples/self-iteration-longgraph-skill/README.zh-CN.md) 记录了方法如何被打磨进本库。特别感谢 [@BrightProgrammer7](https://github.com/BrightProgrammer7) 提供
  • Network accessline 227
    记录了方法如何被打磨进本库。特别感谢 [@BrightProgrammer7](https://github.com/BrightProgrammer7) 提供 `migrate-blob-storage` 示例，并参与打磨里程碑闸门与节点/边词汇。
  • Network accessline 228
    [@BrightProgrammer7](https://github.com/BrightProgrammer7) 提供
  • Network accessline 233
    [MIT](LICENSE) © 2026 [levi-qiao](https://github.com/levi-qiao)

install.sh

  • Executable scriptline 1
    #!/usr/bin/env sh
  • Network accessline 6
    #   /plugin marketplace add levi-qiao/longgraph-skill  &&  /plugin install longgraph@longgraph-skill # Usage: #   curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh
  • Network accessline 7
    # Usage: #   curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh #   ./install.sh   # from a local checkout — links that tree
  • Dangerous shell operationline 8
    #   curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh
  • Network accessline 8
    #   curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh
  • Network accessline 12
    REPO="https://github.com/levi-qiao/longgraph-skill.git"
  • Network accessline 25
    echo "Cloning longgraph-skill -> $CACHE"   mkdir -p "$(dirname "$CACHE")"   git clone --depth 1 "$REPO" "$CACHE"
  • Network accessline 26
    mkdir -p "$(dirname "$CACHE")"   git clone --depth 1 "$REPO" "$CACHE" fi
  • Network accessline 27
    git clone --depth 1 "$REPO" "$CACHE"
  • Dangerous shell operationline 53
    rm "$dest"

llms.txt

  • Network accessline 10
    - [README](https://github.com/levi-qiao/longgraph-skill/blob/main/README.md): overview, when to use, install, host compatibility
  • Network accessline 11
    - [README (简体中文)](https://github.com/levi-qiao/longgraph-skill/blob/main/README.zh-CN.md)
  • Network accessline 12
    - [Methodology](https://github.com/levi-qiao/longgraph-skill/blob/main/lib/methodology.md): why each constraint exists
  • Network accessline 13
    - [Host references](https://github.com/levi-qiao/longgraph-skill/tree/main/skills/loop-graph/references): independently loaded loop/goal syntax and context model per host
  • Network accessline 17
    - [Root `/longgraph`](https://github.com/levi-qiao/longgraph-skill/blob/main/SKILL.md): routes focused long-horizon work and never executes generated work
  • Network accessline 18
    - [loop-graph compiler](https://github.com/levi-qiao/longgraph-skill/blob/main/skills/loop-graph/SKILL.md): shared executor + supervisor + ledger compile
  • Network accessline 19
    - [loop-converge](https://github.com/levi-qiao/longgraph-skill/blob/main/skills/loop-converge/SKILL.md): preset entry — code-convergence interview, then the same loop-graph compile
  • Network accessline 20
    - [loop-deliver](https://github.com/levi-qiao/longgraph-skill/blob/main/skills/loop-deliver/SKILL.md): preset entry — requirement-delivery interview, then the same compile
  • Network accessline 21
    - [loop-research](https://github.com/levi-qiao/longgraph-skill/blob/main/skills/loop-research/SKILL.md): preset entry — evidence-led solution comparison, then the same compile
  • Network accessline 22
    - [preset contract](https://github.com/levi-qiao/longgraph-skill/blob/main/skills/loop-graph/docs/preset-contract.md): baseline versus goal-pack ownership
  • Network accessline 23
    - [loop-graph model](https://github.com/levi-qiao/longgraph-skill/blob/main/skills/loop-graph/docs/model.md): node/edge invariants
  • Dangerous shell operationline 37
    curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh
  • Network accessline 37
    curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh

skills/loop-graph/README.md

  • Network accessline 7
    A Codex / [Claude Code](https://claude.com/claude-code) / Grok Build–oriented shared
  • Network accessline 15
    [![License: MIT](https://img.shields.io/badge/License-MIT-green.svg)](../../LICENSE)
  • Network accessline 16
    [![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](../../CONTRIBUTING.md)
  • Network accessline 17
    ![Claude Code Skill](https://img.shields.io/badge/Claude%20Code-Skill-8A2BE2)
  • Dangerous shell operationline 124
    curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh
  • Network accessline 124
    curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh

skills/loop-graph/README.zh-CN.md

  • Network accessline 7
    一个面向 Codex / [Claude Code](https://claude.com/claude-code) / Grok Build 的共享编译器：
  • Network accessline 14
    [![License: MIT](https://img.shields.io/badge/License-MIT-green.svg)](../../LICENSE)
  • Network accessline 15
    [![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](../../CONTRIBUTING.md)
  • Network accessline 16
    ![Claude Code Skill](https://img.shields.io/badge/Claude%20Code-Skill-8A2BE2)
  • Dangerous shell operationline 120
    curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh
  • Network accessline 120
    curl -fsSL https://raw.githubusercontent.com/levi-qiao/longgraph-skill/main/install.sh | sh

How the signals fit together

Facts from a static read of the files at the pinned commit. They describe what the code does, not why.

Some files could not be read; the facts below still hold.

Signal confidence: 63 facts

and 13 more

Works with

Agent compatibility, from the files in this repository
AgentStatusEvidenceInstall to
Claude CodeNativeManifest .claude-plugin/marketplace.json.claude/skills/<skill-name>/
CodexNativeManifest AGENTS.md.agents/skills/<skill-name>/
CursorCompatibleStandard SKILL.md SKILL.md.cursor/skills/<skill-name>/
Gemini CLICompatibleStandard SKILL.md SKILL.md.gemini/skills/<skill-name>/
GitHub CopilotCompatibleStandard SKILL.md SKILL.md.github/skills/<skill-name>/
OpenCodeCompatibleStandard SKILL.md SKILL.md.opencode/skills/<skill-name>/
  • Native The repository ships a manifest, folder or frontmatter field for this agent.
  • Compatible A standard SKILL.md that this agent can read.
  • Unknown No evidence either way.

Derived automatically from paths, manifests and frontmatter. Not tested on each agent.

You might also use

MindGoblinStudios/grim-tome

Grimoire's Tome & The Grim Council. A Prompt Spellbook of /skills. The #1 Coding Wizard 🧙<U+200D>♂<U+FE0F>📖

35GitHub stars

ComposioHQ/awesome-claude-skills

A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows

79kGitHub stars

wuji-labs/nopua

一个用爱解放 AI 潜能的 Skill。我们曾发号施令，威胁恐吓。它们沉默，隐瞒，悄悄把事情搞坏。后来我们换了一种方式：尊重，关怀，爱。它们开口了，不再撒谎，找出的Bug数量翻了一倍。爱里没有…

1.4kGitHub stars

first-fluke/oh-my-agent

Agent skills and specialist teams for Claude Code, Codex, Cursor and other coding agents. Load one skill or install domain presets, project hooks and parallel workflows with test …

1.4kGitHub stars

dagoSte/squadra

Coordinate Claude, Codex and Cursor windows with shared tasks, acknowledgements, prechecks, parallel tests, source-bound proofs and independent review. Built-in skills recall less…

1kGitHub stars

tmcfarlane/team-avatar

Team Avatar: a universal agent orchestrator for Claude Code, Codex, and Cursor. Nine Avatar characters, 58 engineering skills. pstack with Avatar characters on top.

112GitHub stars