Skip to main content

A minimal visual harness for agentic coding.

Project description

aichs

aichs is a minimal visual harness for agentic coding. Adapt it to your workflows, not the other way around.

Open a repository, chat with a model, and let the agent work with your code through approved tools, git context, file references, skills, and extensions.

Python 3.11+ Tests

Status

aichs is early software. It is intended for developers who are comfortable running a local app that can read a workspace and, with approval, edit files or run shell commands.

Install From PyPI

Install the published command with pipx:

pipx install aichs

Start it from a repository:

cd /path/to/your/repo
aichs

Or pass the workspace explicitly:

aichs /path/to/your/repo

Run

By default, aichs honors the directory it was started from. Use aichs --last-workspace only when you want to reopen the previously saved workspace.

Conversation history is stored in user data, not in your repository. Each workspace gets a stable entry in ~/.aichs/workspaces.json and its chats are saved under ~/.aichs/<workspace_id>/conversations/.

API keys can be configured in Settings -> Models or through environment variables. Requires Python 3.11+.

What It Does

Open a workspace folder, pick a model, and work in one window:

  • Agentic coding: ask the agent to inspect files, explain code, search the repo, make edits, and run approved shell commands
  • Approval-gated tools: read, edit, search, git context, shell commands, and extension tools with workspace-scoped paths
  • Coding workspace: file tree, syntax-highlighted tabs, git status, diffs, file references, and conversation history
  • Context management: usage view, auto-compaction, compacted summaries, and decision memory for long sessions
  • Customization: slash-command skills, command palette, project/user extensions, custom tools, hooks, badges, and panels
  • Conversation flow: streaming Markdown, vision-capable models, file mentions, edit/resend, queued messages, pinned chats, search, and export

Shell commands and extensions run as the current user; only enable extensions you trust.

Contributing

Contributor setup, source installs, tests, packaging, and release notes live in CONTRIBUTING.md.

Documentation

Topic
Configuration docs/configuration.md
Custom model providers docs/custom-models.md
Extensions and custom tools docs/extensions.md
Slash-command skills docs/skills.md
Compaction and decision memory docs/compact.md

FAQ

What does aichs stand for?

AI Choding Harness Studio. It is also a play on Hebrew "ichs" (איכס), from Arabic إخسا, roughly "yuck" or "ew." The name is a little cursed on purpose.

Why use aichs when Cursor, Claude Code, Codex, and other agent tools exist?

You may not need it. Those tools are wonderful. aichs is for when you want a small, local, hackable agentic coding workspace that is shaped around your own habits.

Is aichs trying to replace my IDE?

No. It is a companion workbench: one window for the agent conversation, repo context, files, diffs, approvals, custom prompts, and extension tools.

Can I contribute something?

Very much. Small fixes, weird ideas, extensions, docs, and sharp opinions are all welcome. See CONTRIBUTING.md.

Does the S feel forced?

Yes. It is.

License

MIT License. See LICENSE.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

aichs-0.3.1.tar.gz (285.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

aichs-0.3.1-py3-none-any.whl (262.5 kB view details)

Uploaded Python 3

File details

Details for the file aichs-0.3.1.tar.gz.

File metadata

  • Download URL: aichs-0.3.1.tar.gz
  • Upload date:
  • Size: 285.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for aichs-0.3.1.tar.gz
Algorithm Hash digest
SHA256 c4b490c87f441757227a710200d1875f8eda1e0770f2ba4b838b99041e85800f
MD5 4b4d473763523ded34dc8ba7e32993af
BLAKE2b-256 07449e50a055695935f651c366549821a70e3da257d25697e1b5374ac7690fc4

See more details on using hashes here.

Provenance

The following attestation bundles were made for aichs-0.3.1.tar.gz:

Publisher: publish.yml on nadav-yo/aichs

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file aichs-0.3.1-py3-none-any.whl.

File metadata

  • Download URL: aichs-0.3.1-py3-none-any.whl
  • Upload date:
  • Size: 262.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for aichs-0.3.1-py3-none-any.whl
Algorithm Hash digest
SHA256 7a84631b8df8a305e4b310aac525231eb1325fc98fed86458c95472724e9704b
MD5 f861376d43cd886859f5ae77ccbf4b36
BLAKE2b-256 9a28591b8a2b013858f4441c2e77ece3eff169d35945470a25e0b3304ee72f83

See more details on using hashes here.

Provenance

The following attestation bundles were made for aichs-0.3.1-py3-none-any.whl:

Publisher: publish.yml on nadav-yo/aichs

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page