/jump to repo

Repos Recent Bookmarks Watched Notes Tags Discover Compare Stats GitHub

Repository brief

shareAI-lab/learn-claude-code

Read the upstream summary on the left, browse the cached forks below it, and load each fork comparison into the right-hand panel.

Cached analysis

cached 2026-03-30T19:51:10.861Z

1mo ago

shareAI-lab/learn-claude-code

shareAI-lab/learn-claude-code is an active Python-based open source project for a "nano claude code-like agent harness" focused on harness engineering for real agents. It has a documentation site, multilingual README files, and a sizable community footprint with 43,808 stars and 6,704 forks. The recent commit history shows active maintenance around agent/session behavior, compaction, task graph logic, and web UI syncing.

Loading tags...

Stars43,808

Forks6,704

Default branchmain

Last pushed2026-03-29T16:09:39Z

Recommended shortcuts

Jump straight into Discofork's strongest cached fork picks, or open a compare view in one click.

Forks

Choose a fork to inspect

10 of 10 fork briefs

Maintenance:

Magnitude:

Sort:

Selected

Choose this fork if you want a Java/Spring AI rewrite with extra agent/team/worktree workflows and you are comfortable owning divergence from upstream. Stay with upstream if you want the latest harness fixes, broader documentation coverage, and the original Python-first implementation.

Choose this fork only if you specifically want an older snapshot of learn-claude-code. For active use, upstream is the better choice because this fork has no visible enhancements and is 54 commits behind.

Choose this fork only if you specifically want a near-vanilla snapshot for local experimentation; otherwise prefer upstream, since it is materially newer and this fork shows no compensating additions.

Choose this fork only if you want a near-vanilla snapshot and are comfortable missing recent upstream fixes; otherwise the upstream repo is the better choice for active use.

Choose this fork if your goal is reverse engineering, architecture study, or rebuilding Claude Code behavior from analysis materials. Choose upstream if you want the actively maintained Python harness with recent session, compaction, task-graph, and web UI fixes.

Choose this fork if your goal is analysis, reconstruction, or learning from Claude Code internals. Choose upstream if you want an actively maintained agent harness with recent fixes and a more complete product workflow.

Prefer this fork if you want a tutorial-focused derivative with a travel-assistant workflow and are comfortable losing a lot of upstream documentation. Prefer upstream if you want the most complete, actively maintained harness with recent fixes and the full multilingual learning material.

Choose this fork if your goal is analysis, documentation, or reimplementation research. Choose upstream if you want a current, maintained Claude Code-like harness with working agent-session behavior and recent fixes.

Choose this fork if your priority is backend flexibility and OpenAI-based deployment. Choose upstream if you want the latest harness fixes, upstream docs, and the most actively maintained canonical line.

Prefer upstream unless you specifically need this fork as a frozen starting point; it adds no new capabilities and is materially behind on fixes.

Fork comparison

abel533/learn-claude-code

70/100

active

significant_divergence

Choose this fork if you want a Java/Spring AI rewrite with extra agent/team/worktree workflows and you are comfortable owning divergence from upstream. Stay with upstream if you want the latest harness fixes, broader documentation coverage, and the original Python-first implementation.

Likely purpose

To provide a Spring AI-based alternative for users who want the learn-claude-code harness concepts in a Java ecosystem, with added agent/team orchestration and worktree management.

Best for

Java or Spring developers who want a Claude Code-like harness without adopting the upstream Python stack.; Teams experimenting with multi-agent coordination, task protocols, and worktree-based agent workflows.; Adopters who value a local architectural rewrite over staying close to upstream.; People willing to trade upstream parity for an ecosystem fit in Spring AI.

Additional features

Spring AI-based rewrite in Java instead of the upstream Python stack.
New autonomous-agent and team orchestration code paths, including team protocols and teammate management.
Worktree management support, which suggests stronger multi-branch or multi-agent workflow handling.
Documentation and version metadata updates tailored to the fork’s own Spring AI implementation.
Cleaner operational logging around todo updates, with debug output removed.
Expands API, schema, or SDK surface area for downstream integrations and generated clients.

Missing features

The upstream Python-based implementation and its direct dependency stack (`anthropic`, `python-dotenv`, `pyyaml`) are not what this fork is centered on.
It is 51 commits behind upstream, so it likely lags recent upstream fixes around compaction, task-claim races, output formatting, and web UI sync behavior.
The fork shows a large rewrite of `README-zh.md`, which suggests some upstream documentation coverage may have been reduced or reworked.
Upstream’s multilingual, actively maintained documentation and recent harness fixes may be harder to use here until the fork catches up.
Some upstream behavior appears to be removed, disabled, or intentionally stripped down in this fork.
It trails upstream by 51 commits, so some recent upstream features and fixes are likely not present yet.

Strengths

Good fit if you want the same project idea in a Java/Spring AI environment.
Adds workflow-oriented code for coordinating multiple agents and managing worktrees.
Recent commits indicate active local development rather than a dead fork.
Likely easier to adopt for teams already standardized on Spring and Java.

Risks

Materially diverged from upstream, so upstream docs, examples, and bugfixes may not transfer cleanly.
Being behind upstream means you may miss important harness reliability fixes.
A Spring port can be harder to merge with upstream changes over time.
The fork’s narrower ecosystem focus may reduce community support compared with upstream.