← Open Source
codejunkie99

fable-orchestrator

Fable 5.1 orchestrates. GPT-5.6 Luna and DeepSeek V4 Flash implement.

AI EngineeringGive agent toolsBuild agent workflow (SDK)Shell
Open on GitHub
Momentum
+0stars in 24 hours0.0%
615
Stars
110
Forks
+1
This week
0
Contributors
Created 2026-09-02 · Updated 2026-10-15 · #16042 today

codejunkie99/fable-orchestrator is ranked 16,042 with 615 stars.

Same-category comparison

codejunkie99/fable-orchestrator is ranked 16,042 among these 8 peers; aperoc/toolkami is ranked 16,041.

Same-category comparison
NameRankMetrics
ynulihao/AgentSkillOS16,035
Language:
HTML
Stars:
616
Forks:
77
Description:
Build your agent from 200,000+ skills via skill RETRIEVAL & ORCHESTRATION.
RoboStack/jupyter-ros16,038
Language:
Python
Stars:
615
Forks:
114
Description:
Jupyter widget helpers for ROS, the Robot Operating System.
ankane/informers16,040
Language:
Ruby
Stars:
615
Forks:
17
Description:
Fast transformer inference for Ruby.
aperoc/toolkami16,041
Language:
Ruby
Stars:
615
Forks:
32
Description:
Simple Agents Made Easy.
hald/things-mcp16,043
Language:
Python
Stars:
615
Forks:
68
Description:
Things.app MCP Server.
leodavinci1/kanbots16,044
Language:
TypeScript
Stars:
615
Forks:
55
Description:
Local collaboration interface for working on a kanban board where each task is either a Claude Code or Codex agent.
Denario-AI/Denario16,051
Language:
TeX
Stars:
614
Forks:
92
Description:
Modular Multi-Agent System for Scientific Research Assistance.
dair-ai/dair-academy-plugins16,054
Language:
HTML
Stars:
614
Forks:
64
Description:
DAIR.AI Academy Plugins.
Key parameters and specifications
Key parameters and specifications
ParameterValue
LanguageShell
CategoryAI Engineering
SubcategoryGive agent tools, Build agent workflow (SDK)
KeywordsAgent-ready APIs, Multi-model workflow, Planning, Skills
Stars615
Forks110
Contributors0
Created date2026-09-02
Updated date2026-10-15
Stars in 1 day0
Stars in 7 days1
Rank16,042
Previous rank15,842
Rank change-200
7-day stars rank11,155
Previous 7-day stars rank11,209
7-day stars rank change54
Total stars rank16,461
Previous total stars rank16,436
Total stars rank change-25
README summary

Fable orchestrator Fable is a small, local-first routing skill for Codex. 1 plans and adjudicates; it does not write code or own the workspace. Codex remains the runtime and delegates bounded implementation work to OpenCode Go agents: GPT-5. DeepSeek V4 Flash handles loops, repeated iteration, and high-throughput implementation. 1 remains outside the implementation graph for planning and final adjudication. The skill consumes the callable and agents supplied by Codex Router. It does not ship a proxy, dashboard

, model catalog, credential store, or API key. Configure OpenCode Go once in the router, then restart Codex after changing provider or agent definitions. Repository layout The three files under are the installable skill. is executable and invokes Claude Code's local alias with no session persistence. The packet passed to it must contain decisions and workspace facts only; never put credentials in a packet. Install From this repository: prints the exact destination and copy operations without creating files. installs to , creates

only the required directories, and is safe to run again. Use to select another skills directory: The installer reads only this repository and the destination path. It never reads, creates, or modifies credentials. Start a new Codex task after changing the provider or agent definitions. Use Invoke the skill with an objective: Fable returns a bounded graph. Codex validates the graph, starts ready workers in parallel when useful, collects their evidence, verifies the result, and asks Fable to adjudicate when

the task needs another decision. Every implementation node must use GPT-5. If neither allowed OpenCode Go route is callable, the workflow reports the blocker instead of inventing a model. Test Run the repository's dependency-light checks: The test checks shell syntax, the copied source files, required routing strings, basic YAML structure, optional XML validation, SVG safety constraints, a temporary-home dry run, an idempotent copy, and common credential-shaped strings. It does not require PyYAML or a live Claude

login.

Frequently asked questions

How many stars and forks are recorded?

codejunkie99/fable-orchestrator has 615 stars and 110 forks.

What is the repository rank?

codejunkie99/fable-orchestrator is ranked 16,042.

What language and category are recorded?

codejunkie99/fable-orchestrator has 615 stars, uses Shell and belongs to AI Engineering.

Which repository is closest in rank?

codejunkie99/fable-orchestrator is compared with aperoc/toolkami, ranked 16,041 with 615 stars.

README

Fable orchestrator

Fable is a small, local-first routing skill for Codex. Claude Fable 5.1 plans and adjudicates; it does not write code or own the workspace. Codex remains the runtime and delegates bounded implementation work to OpenCode Go agents:

  • GPT-5.6 Luna handles normal implementation.
  • DeepSeek V4 Flash handles loops, repeated iteration, and high-throughput implementation.
  • Fable 5.1 remains outside the implementation graph for planning and final adjudication.

The skill consumes the callable opencode-go/ and opencode-go-responses/ agents supplied by Codex Router. It does not ship a proxy, dashboard, model catalog, credential store, or API key. Configure OpenCode Go once in the router, then restart Codex after changing provider or agent definitions.

Repository layout

skill/fable/
├── SKILL.md
├── agents/openai.yaml
└── scripts/ask_fable.sh
assets/fable-orchestrator.svg
install.sh
tests/test_skill.sh

The three files under skill/fable/ are the installable skill. ask_fable.sh is executable and invokes Claude Code's local fable alias with no session persistence. The packet passed to it must contain decisions and workspace facts only; never put credentials in a packet.

Install

From this repository:

./install.sh --dry-run
./install.sh --copy

--dry-run prints the exact destination and copy operations without creating files. --copy installs to ~/.codex/skills/fable, creates only the required directories, and is safe to run again. Use --target DIR to select another skills directory:

./install.sh --copy --target "$PWD/.local/codex/skills"

The installer reads only this repository and the destination path. It never reads, creates, or modifies credentials. Start a new Codex task after changing the provider or agent definitions.

Use

Invoke the skill with an objective:

$fable build the feature

Fable returns a bounded graph. Codex validates the graph, starts ready workers in parallel when useful, collects their evidence, verifies the result, and asks Fable to adjudicate when the task needs another decision. Every implementation node must use GPT-5.6 Luna or DeepSeek V4 Flash. If neither allowed OpenCode Go route is callable, the workflow reports the blocker instead of inventing a model.

Test

Run the repository's dependency-light checks:

tests/test_skill.sh

The test checks shell syntax, the copied source files, required routing strings, basic YAML structure, optional xmllint XML validation, SVG safety constraints, a temporary-home dry run, an idempotent copy, and common credential-shaped strings. It does not require PyYAML or a live Claude login.

License

MIT. See LICENSE.

Nearby in the ranking
Similar items