by GanyuanRan
AI coding with Aegis governance built into execution: baseline-aware changes, evidence-backed delivery. Free desktop client, your choice of model. 将哲科思维融入 AI 开发执行，让变更有依据、交付有证据。用创意编织现实。
# Add to your Claude Code skills
git clone https://github.com/GanyuanRan/AutoloomSee how Autoloom compares with popular alternatives.
Autoloom is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by GanyuanRan. AI coding with Aegis governance built into execution: baseline-aware changes, evidence-backed delivery. Free desktop client, your choice of model. 将哲科思维融入 AI 开发执行，让变更有依据、交付有证据。用创意编织现实。. It has 73 GitHub stars.
Autoloom's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
Clone the repository with "git clone https://github.com/GanyuanRan/Autoloom" and add it to your Claude Code skills directory (see the Installation section above).
Autoloom is primarily written in JavaScript. It is open-source under GanyuanRan on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh Autoloom against similar tools.
No comments yet. Be the first to share your thoughts!
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
AI coding with engineering judgment built into execution.
Autoloom brings Aegis’s core governance methods into the coding runtime: question whether a change is needed, challenge assumptions before editing, and review delivery against requirements and execution evidence.
Spend less time reminding your agent to investigate first, avoid unnecessary complexity, and show what it actually verified.
| Method | Its role in the development workflow |
|---|---|
| First principles and change necessity | At task initiation, examine goals, facts, existing capabilities, and constraints to establish why a change is needed and whether an adequate solution already exists. |
| Anti-entropy and structural complexity checks | Before changes, examine duplicated responsibilities, unnecessary layers, and maintenance burden so added complexity addresses a concrete problem. |
| Counterexample review and causal diagnosis | Before changes, examine conditions that could invalidate a solution; after execution failures, require diagnosis grounded in observations and evidence. |
| Implementation conformance and completion review | Before recording results, compare task requirements, actual execution, and verification material, stating what is complete and what remains unverified. |
Autoloom delivers these method requirements at the relevant points, checks prerequisite ordering for key operations, and retains process records. Whether a method was applied well and a result is reliable still needs to be judged from code, execution results, and verification evidence.
Recorded project goals, constraints, and key decisions stay with the project. Changes to those decisions present a specific diff for your review. You can also configure a test-first preference and code complexity thresholds to reflect your project's requirements.
Aegis is our open-source AI engineering method pack. Autoloom integrates its core governance methods into development, covering change necessity, complexity, causal diagnosis, and delivery review.
Guidance loads as needed. The runtime maintains prerequisites for key actions, task state, and execution evidence—connecting governance to actual work and reducing repeated reminders and manual checks. Reliable delivery still depends on actual verification results.
Autoloom is built on DeepSeek's official open-source DeepSeek Harness dsh-v0.1.1-rc.2, at source commit b150a551, retaining the Cordis plugin mechanism.
We will continue following official upstream releases and learning from worthwhile improvements. Changes that fit Autoloom and solve real problems will be adopted after adaptation and verification; selected later reliability fixes are already included.
We also maintain our own product direction: bring engineering judgment into automated development, ground the process in evidence, make results inspectable and keep reducing the user's supervision burden. Autoloom has its own version numbers and release cadence.
We thank DeepSeek Harness, Cordis and the related open-source projects for their contributions. The client retains third-party licenses and copyright notices.
Become a trustworthy, fully automated AI software development platform.
Help ideas progress through requirements, design, implementation, verification, and continued iteration. Make more of that work autonomous while keeping key decisions, supporting evidence, and delivered results inspectable.
That ambition comes with an ongoing question: how can we achieve the best possible balance of cost, speed, and quality?
We care about the total cost of reliably delivering a requirement: model calls, waiting time, human supervision, and rework caused by defects. Reducing unproductive attempts and repeated context, and directing verification effort toward what matters to the result, are continuing priorities. Different tasks need different tradeoffs. We aim to improve how much of all three can be achieved together and test those improvements on real tasks.
Autoloom is in Alpha. Fully automated development and a better balance across these three dimensions are goals we are working toward.
macOS and Linux desktop clients are planned, so you can use Autoloom on your preferred operating system. Windows x64 is currently available; progress on macOS and Linux will be announced in this repository.
If you want AI development to include this kind of engineering judgment, ⭐ Star Autoloom to support its continued development. Follow the link to the top of the repository, then click Star in the upper-right corner.
For new releases and repository activity, select Watch → All Activity at the top right of this repository. You can adjust notification delivery in GitHub settings. Use Star to bookmark and show support; use Watch to subscribe to notifications.
Autoloom-<version>-win-x64.exe.The client is free; your chosen provider bills model usage.
The workspace supports local projects and Linux x64 projects connected over SSH, with project files, code diffs, command logs and previews. Remote environment requirements are listed in the Alpha notes below.
The installed client supports background checks and downloads for Alpha updates. Once a download is ready, click “Restart to update” to install and reopen the client. Finish or stop active tasks first; ordinary exit does not install an update. Users of the older extracted ZIP must install the first installer release manually.
The user documentation covers model providers, permissions, local and SSH projects, Session backup, updates, and troubleshooting.
Tell us through Issues what Autoloom helped you finish, which steps still needed repeated reminders, and where you hesitated to accept its results. Your actual experience will help us choose what is most worth improving next.
When reporting a problem, include the version, system information, reproduction steps, and expected and actual results. Remove keys, access tokens and personal information before sharing logs or screenshots.
This repository is Autoloom's official hub for downloads, release notes and feedback.
Autoloom’s client source code is not currently public. As the project evolves, we’ll evaluate what to open source and how, taking community feedback and sustainable maintenance into account. The current closed-source approach is not a commitment to staying closed source permanently. We’ll share concrete plans in this repository when they’re ready.
SHA256SUMS. Check the file with PowerShell using Get-FileHash -Algorithm SHA256 <installer-path>.