Skip to main content

see-your-agents

A local dashboard for Claude Code and Codex CLI sessions: who is working, who is waiting for you, which ticket a session belongs to, what was done and where the agent is in its plan. It reads what the programs already keep on disk, so it works with any session, including plain claude or codex in any terminal. From the page you can also write to sessions, answer Claude Code's permission prompts, start new sessions and talk to a supervisor session that watches all the others. The dashboard runs on your machine: a Python server on 127.0.0.1 and a page in your browser.

The board: sessions in the columns Needs you, Working, Your move and Recently closed, with the conversation list on the left

Try the demo in your browser: made-up sessions, nothing to install.

Features

  • Board: sessions in columns by what is needed from you ("Needs you", "Working", "Your move"), with tickets, tasks, cost or tokens, context fill and MR state; the same sessions in an office view.
  • Conversation: read any session as a chat and write to it, whether it runs in VS Code, tmux or the background, or is already closed; group chats with several agents at once.
  • Permissions: answer Claude Code's permission prompts, questions and plans from the page.
  • New session and Fork: launch claude --bg or codex exec in a project folder, or fork a session.
  • Supervisor: a Claude Code session that watches the others over the dashboard's MCP server and hands out tasks, with your confirmation.
  • To-do, search, tools (skills, MCP servers, hooks and how often agents use them), keyboard shortcuts.
  • Builder: ask the supervisor for a new module, and an agent builds it in a sandboxed branch for you to accept.
  • Six UI languages: English, Russian, Spanish, German, French, Simplified Chinese. All UI text was written by Claude Opus 5.5; if a translation or a name could be better, suggestions are welcome.

A permission prompt answered from the board; the terminal's dialog stays open, and the first answer wins:

Answer on a card opens the request with its command; Allow sends it, and the card moves to Working

Quick start

uvx see-your-agents serve           # try it without installing

uv tool install see-your-agents     # or: pipx install see-your-agents
sya serve                           # http://127.0.0.1:8765, opens a browser tab
sya ls                              # the same board in the terminal (--all adds the archive)

From a checkout, for development or for the builder (it works on the dashboard's own git repository):

git clone https://github.com/vldplcd/see-your-agents.git && cd see-your-agents
uv run sya serve

The page is in English by default; Appearance → Language switches it. Optional one-time setup (the supervisor, the permission hook, the VS Code bridge) is in docs/configuration.md.

Requirements

  • uv and Python 3.11 or newer.
  • The Claude Code CLI (claude) in PATH; the Codex CLI (codex) to see its sessions too.
  • macOS or Linux. Windows is not supported natively; under WSL it should work as on Linux (not tested yet).
  • macOS only: the builder (its sandboxes use sandbox-exec) and keeping the Mac awake.
  • Optional: glab for MR states, VS Code or tmux for typing into a live session's terminal, vplan for visual plans.

The dashboard writes nothing to ~/.claude or ~/.codex itself, except through the setup commands you run. Details: docs/configuration.md.

Documentation

Contributing

Contributions are welcome. Ways to help:

  • Bug reports: an issue with your OS, Python, the CLI versions (claude --version, codex --version), the steps and the logs.
  • Feature modules: a new view, panel, card line or MCP tool.
  • Providers: support for another agent CLI next to Claude Code and Codex.
  • Translations: a new UI language, or better wording in an existing one.
  • Docs: corrections, clearer explanations, screenshots.

How a feature is added: a folder in src/see_your_agents/features/<id>/ and one line in src/see_your_agents/features/__init__.py; the host and other features are not edited. Walkthrough: docs/adding-a-feature.md. A provider is the same shape, a folder in core/providers/<id>/ and one registry line: docs/providers.md.

The rules, in short:

  • Work on a branch of your fork; one logical change per commit, subject Area: what changed (todo, core, app, docs, ci…), the body says why.
  • uv run ruff check src tests && uv run ruff format --check src tests && uv run pytest -q passes before you open a pull request; CI runs it on Linux and macOS.
  • Every UI string is a catalog key with an English string; agent-facing text is English.
  • UI changes come with screenshots in light and dark; security-sensitive changes come with tests.

Pull requests written with Claude Code, Codex or another agent are welcome; the repo carries rules for them (CLAUDE.md, AGENTS.md). The maintainers may decline any pull request without giving a reason.

Details: CONTRIBUTING.md. Keep it civil (code of conduct), and report vulnerabilities privately as SECURITY.md describes.

License

MIT, see LICENSE.

Metadata

Release files for see-your-agents 0.1.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for see-your-agents 0.1.1
File Size Uploaded Provenance Actions
see_your_agents-0.1.1.tar.gz 647.6 kB With provenance Details for see_your_agents-0.1.1.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for see-your-agents 0.1.1
File Interpreter ABI Platform Actions
see_your_agents-0.1.1-py3-none-any.whl Python 3 none any Details for see_your_agents-0.1.1-py3-none-any.whl

Total release size: 1.1 MB

Release files / see_your_agents-0.1.1.tar.gz

Download URL see_your_agents-0.1.1.tar.gz
Size 647.6 kB
Tags Source
SHA-256 checksum
How to use checksums
e9811143348d17753c78fcd87c02bdc98a2f8ad3261c273b419e878676fd88b2
BLAKE2b-256 checksum
How to use checksums
0d378a52be2db1dda494e266f6550dc659c06085484cecd25d2811ba69b90747
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 4, 2026.

Transparency log

Release files / see_your_agents-0.1.1-py3-none-any.whl

Download URL see_your_agents-0.1.1-py3-none-any.whl
Size 492.8 kB
Tags Python 3
SHA-256 checksum
How to use checksums
3068bb827e8f4bcec97bb063c0e81790991ed277bbcd744e194fef759daa0b42
BLAKE2b-256 checksum
How to use checksums
4009a8090d438ff959de32d6dfadd79d661b7ce9c21c8a2e05d862e05abbdbd9
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.16

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 4, 2026.

Transparency log

Security for see-your-agents 0.1.1

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Full provenance

All 2 files in this release have been uploaded with provenance metadata.

Select from 2 files

Attestation for see_your_agents-0.1.1.tar.gz

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Nov 4, 2026.

Transparency log

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page