Skip to content
View qte77's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report qte77

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
qte77/README.md

qte77

qte77 is a polyrepo orchestration framework for AI coding agents — coordinating Claude Code and compatible agents across 30+ repositories so goals, specs, builds, and learnings compound instead of drift. Agents drive it; humans approve and steer.

What

qte77 is the META layer that keeps 30+ agent-driven repos coherent — a shared map so goals, specs, builds, and learnings compound instead of drift.

  • Agents — Claude Code and compatible LLM coding agents running per-repo, coordinated by orchestrators in this workspace.
  • Orchestratorspolyforge-orchestrator drives the dev loop; office-forge-orchestrator the office loop.
  • Engines — reusable components orchestrators compose: doc-pipeline-engine, polyfetch-scrape.
  • Humans — approve goals, review PRs, steer the orchestrators. Agents propose; humans decide.

qte77 Mental Model — clusters, flow, feedback loop

How

Each repo carries its own quickstart. Start with an orchestrator:

30+ companion repos live as siblings under qte77.

Why

Agentic development across 30+ repos drifts without a shared map — learnings don't flow back into specs, and the system forgets. qte77 fixes that feedback loop (goals → specs → builds → learnings → goals) so the system compounds instead.

It's a cross-repo coordination layer, not a single-repo agent runner or per-prompt orchestrator. Reach for something else if you're building one agent for one repo, or your loop fits in a single prompt.

Refs

License

Apache-2.0 — see LICENSE.

Tools

PythonTypeScriptLinuxBashGitGitHub ActionsDockerDev ContainersAzureAWSGoogle CloudROSHugging FaceGitHub ModelsClaude

Posts

Profile

More: Topics, Interests, TODO.

Lineage

How the current system got here — proof of work, not required reading.

context-engineering-template-legacy (first commit 2025-07-07) coined the BRD → PRD → FRD pipeline and the "CABIO" vision. The work was carried forward in RAPID-spec-forge (Requirements-to-Agent Pipeline & Implementation Driver; formerly CABIO-test) from 2026-02-26, archived 2026-04-26 as RAPID-spec-forge-legacy. Its parts were then decomposed under qte77 — now the META layer for 30+ repos:

Pinned Loading

  1. agentic-market-research-to-gtm agentic-market-research-to-gtm Public

    Stop Guessing. Start Dominating. - Automated market intelligence for AI startups that refuse to fail

    Makefile 2

  2. Agents-eval Agents-eval Public archive

    A Multi-Agent System (MAS) evaluation framework using PydanticAI that generates and evaluates scientific paper reviews through a three-tiered assessment approach: traditional metrics, LLM-as-a-Judg…

    Python 2 1

  3. context-engineering-template-legacy context-engineering-template-legacy Public archive

    This project aims to implement a template for context engineering with coding agents.

    Makefile 11 1

  4. gha-dirtree-to-readme gha-dirtree-to-readme Public

    Copy directory tree into Readme.

    Python

  5. gha-ai-changelog gha-ai-changelog Public

    Update CHANGELOG.md with AI Action

    Makefile

  6. gha-sec-feed gha-sec-feed Public

    Python