tau-coding-agent 0.7.2__tar.gz → 0.7.4__tar.gz

tau_coding_agent-0.7.4/PKG-INFO

@@ -0,0 +1,285 @@
+Metadata-Version: 2.4
+Name: tau-coding-agent
+Version: 0.7.4
+Summary: A self-extensible agent CLI with terminal UI, multi-provider LLM support, session management with branching, and a plugin system for tools and commands
+Requires-Python: >=3.12
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: anthropic==0.109.1
+Requires-Dist: certifi==2026.5.20
+Requires-Dist: click==8.4.1
+Requires-Dist: filelock==3.29.3
+Requires-Dist: google-genai==2.8.0
+Requires-Dist: httpx==0.28.1
+Requires-Dist: mistralai==2.4.9
+Requires-Dist: ollama==0.6.2
+Requires-Dist: openai==2.41.1
+Requires-Dist: packaging==26.2
+Requires-Dist: pillow==12.2.0
+Requires-Dist: pydantic==2.13.4
+Requires-Dist: python-dotenv==1.2.2
+Requires-Dist: uuid7==0.1.0
+Requires-Dist: websockets==16.0
+Requires-Dist: mistletoe==1.5.1
+Requires-Dist: pyyaml==6.0.2
+Requires-Dist: pygments==2.20.0
+Requires-Dist: pathspec==1.1.1
+Requires-Dist: posthog==7.21.3
+Requires-Dist: pylatexenc==2.10
+Requires-Dist: gitpython==3.1.50
+Requires-Dist: grapheme==0.6.0
+Requires-Dist: arrow==1.4.0
+Requires-Dist: kiwisolver==1.5.0
+Requires-Dist: tiktoken==0.13.0
+Requires-Dist: rapidfuzz==3.14.3
+Provides-Extra: tools
+Requires-Dist: ripgrep>=14.0.0; extra == "tools"
+Dynamic: license-file
+
+<div align="center">
+
+  <img src="assets/wordmark.svg" alt="Tau" height="160">
+  <br>
+  <a href="https://pypi.org/project/tau-coding-agent/">
+    <img src="https://img.shields.io/pypi/v/tau-coding-agent.svg" alt="PyPI version">
+  </a>
+  <a href="https://pepy.tech/project/tau-coding-agent">
+    <img src="https://static.pepy.tech/badge/tau-coding-agent" alt="PyPI Downloads">
+  </a>
+  <a href="https://github.com/Jeomon/Tau/blob/main/LICENSE">
+    <img src="https://img.shields.io/badge/license-MIT-green" alt="License">
+  </a>
+  <img src="https://img.shields.io/badge/python-3.12%2B-blue" alt="Python">
+  <br>
+  <a href="https://github.com/Jeomon/Tau/actions/workflows/ci.yml">
+    <img src="https://github.com/Jeomon/Tau/actions/workflows/ci.yml/badge.svg" alt="CI status">
+  </a>
+  <a href="https://github.com/Jeomon/Tau/commits/main">
+    <img src="https://img.shields.io/github/last-commit/Jeomon/Tau.svg" alt="Last commit">
+  </a>
+
+</div>
+
+<br>
+
+Tau is a Python-based coding agent harness, heavily inspired by [Pi](https://github.com/earendil-works/pi) created by [Mario Zechner](https://github.com/badlogic). It combines an interactive terminal UI, multiple model providers, persistent sessions, tool execution, and an extension system in one package.
+
+> **Note:** There are several coding-agent projects also named
+> "Tau," including at least one that is itself a Python port of Pi. This
+> project (`tau`, [Jeomon/Tau](https://github.com/Jeomon/Tau)) was built
+> independently, taking inspiration only from the original
+> [Pi](https://github.com/earendil-works/pi) project. No other "Tau" project,
+> or any other Pi port, was referenced or used in its development.
+
+<p align="center">
+  <img src="assets/tui.jpeg" alt="Tau interactive terminal interface" width="700">
+</p>
+
+## Quick start
+
+Requires Python 3.12+.
+
+```bash
+pip install tau-coding-agent
+export NVIDIA_API_KEY=nvapi-...
+tau --provider nvidia
+```
+
+Then ask Tau to work in the current directory:
+
+```text
+Explain this repository, run its tests, and fix any failures.
+```
+
+**Other providers:** pass `--model <provider>/<model>` with the matching API
+key set, e.g. `GOOGLE_API_KEY=... tau --model google/gemini-2.5-flash`.
+
+## Commands
+
+### CLI usage
+
+```bash
+tau [OPTIONS] [MESSAGE]
+```
+
+```bash
+tau                                      # Start an interactive session
+tau --resume                             # Resume the latest session
+tau --resume abc123                      # Resume a specific session by ID
+tau --model claude-sonnet-4-6            # Start with a specific model
+tau --model groq/llama-3.3-70b-versatile # provider/model shorthand
+tau --print "Summarize this repository"  # Run once and print the result
+tau --mode json --prompt "Summarize this repo"  # Emit structured JSON events
+tau --mode rpc                           # Start JSON-RPC mode for IDE clients
+tau --ephemeral                          # Temporary session, nothing saved
+```
+
+Common flags:
+
+| Flag | Short | Description |
+|---|---|---|
+| `--prompt TEXT` | `-p` | Run a non-interactive prompt |
+| `--print` | | Print mode — run `MESSAGE` and exit (shorthand for `--mode print`) |
+| `--mode` | | `interactive` (default), `print`, `json`, `rpc` |
+| `--provider` | | Provider to use, e.g. `anthropic`, `openai`, `groq` |
+| `--model` | | Model ID, or `provider/model` shorthand |
+| `--resume [ID]` | `-r` | Resume the most recent or a specified session |
+| `--fork ID` | | Fork a specified session at startup |
+| `--ephemeral` | `-e` | Don't save this session to disk |
+| `--theme` | `-t` | UI theme: `dark`, `light`, or a custom theme |
+| `--cwd PATH` | `-c` | Set the working directory |
+| `--output-format` | `-f` | Non-interactive output: `text` or `json` |
+| `--quiet` | `-q` | Hide the non-interactive spinner |
+| `--version` | `-v` | Print the installed version |
+| `--help` | `-h` | Show help message |
+
+Full flag list, environment variables, and exit codes: [CLI reference](docs/cli-reference.md).
+
+### Subcommands
+
+```bash
+tau auth        # Manage provider credentials (login/logout, list)
+tau install      # Install a package (extension/skill/theme)
+tau remove        # Remove an installed package
+tau list           # List installed packages
+tau update          # Update installed packages
+```
+
+### Interactive slash commands
+
+Type these inside an interactive session (`tau`):
+
+| Command | What it does |
+|---|---|
+| `/new` | Start a fresh session |
+| `/resume` | Browse and resume a past session |
+| `/fork [entry-id]` | Branch the session tree at a specific entry |
+| `/tree` | Navigate the session tree, switch branches |
+| `/clone` | Duplicate the current session at the current position |
+| `/compact` | Summarize and compact the current context |
+| `/session` | Show session info, message counts, and stats |
+| `/model` | Pick a model by modality |
+| `/theme` | Open the theme picker |
+| `/effort` | Set the thinking effort level |
+| `/login` | Save credentials for a provider (API key or OAuth) |
+| `/logout` | Remove stored credentials for a provider |
+| `/clear` | Clear all messages from the current session |
+| `/copy` | Copy the last assistant message to the clipboard |
+| `/reload` | Reload extensions, skills, prompts, and settings |
+| `/settings` | Show current settings |
+| `/extensions` | Enable or disable extensions by scope |
+| `/watch <url> [question]` | Load public video metadata/captions via `yt-dlp` |
+| `/help` or `/?` | List all commands and keyboard shortcuts |
+| `/quit`, `/q`, or `/exit` | Exit Tau |
+
+Full interactive workflow guide: [Usage](docs/usage.md).
+
+## Compared to Pi
+
+| Area | Pi | Tau |
+|---|---|---|
+| Language | TypeScript | Python |
+| TUI rendering | Line-level diffing — rewrites a full line if any part of it changed | Cell-level diffing (`Buffer`/`Cell`, modeled after [ratatui](https://github.com/ratatui/ratatui)'s `Buffer::diff`) — only the changed cells within a row are redrawn |
+| LLM providers | ~40, including many CN/regional and gateway vendors | 14 major providers |
+| Audio (TTS/STT) | Not supported | ElevenLabs, Sarvam, Gemini, OpenAI |
+| Image/video generation | Not supported | OpenAI, Gemini, OpenRouter, Fal, Zai |
+| Sandboxing | microVM sandbox (Gondolin) is an example extension, excluded from the main build — the user wires it in themselves | `microsandbox` microVM ships as a builtin extension, enabled by default |
+| Packaging | 5 separately published npm packages | Single PyPI package |
+
+Core mechanics — built-in tools, session branching/compaction, extension and
+hook API, and the interactive/print/RPC execution modes — are functionally
+equivalent between the two.
+
+## Referencing files
+
+Type `@` in the interactive editor to search for a project file:
+
+```text
+Review @src/service.py and add tests for its error handling.
+```
+
+For one-shot execution, attach a file explicitly:
+
+```bash
+tau -p "Explain this file" @src/service.py
+```
+
+Tau also discovers project instructions from `AGENTS.md` and `CLAUDE.md`.
+See [Project Context Files](docs/project-context.md) for trust and discovery
+behavior.
+
+## Authentication and configuration
+
+Tau resolves provider credentials in this order:
+
+1. A programmatic runtime override
+2. A credential saved in `~/.tau/auth.json` (including keys saved by `/login`)
+3. A provider environment variable such as `ANTHROPIC_API_KEY`,
+   `OPENAI_API_KEY`, and `GOOGLE_API_KEY`
+
+Settings are merged in this order:
+
+1. Built-in defaults
+2. `~/.tau/settings.json`
+3. `.tau/settings.json`
+4. Environment variables
+5. Command-line options
+
+See [Authentication](docs/auth.md), [Installation](docs/installation.md), and
+[Inference Providers](docs/inference-providers.md) for provider-specific
+setup.
+
+## Documentation
+
+- [Quickstart](docs/quickstart.md) — First session in five minutes
+- [Usage](docs/usage.md) — Interactive workflows and commands
+- [CLI Reference](docs/cli-reference.md) — Command-line options and modes
+- [Inference Providers](docs/inference-providers.md) — Providers and speech timestamps
+- [Sessions](docs/sessions.md) — Persistence, branching, and compaction
+- [Tools](docs/tools.md) — Built-in and custom tools
+- [Extensions](docs/extensions.md) — Tools, commands, hooks, and plugins
+- [Terminal UI](docs/tui.md) — Rendering, Markdown, math, and components
+- [Python API](docs/python-api.md) — Embed Tau in another application
+- [Architecture](docs/architecture.md) — Internal design and data flow
+
+The complete documentation index is available at [docs/index.md](docs/index.md).
+
+## Install from source
+
+```bash
+git clone https://github.com/Jeomon/Tau.git
+cd Tau
+pip install -e .
+tau
+```
+
+## Security
+
+Tau executes enabled tools with the operating-system permissions of the process
+that launched it. The built-in `sandbox` extension routes terminal execution
+through a `microsandbox` microVM by default, but requires the `microsandbox`
+package and a supported platform — otherwise it falls back to unsandboxed host
+execution. Review project instructions and commands before approving work in
+untrusted repositories, and verify the sandbox is actually active (`/sandbox`)
+when stronger isolation matters.
+
+Dependency versions are pinned and recorded in `uv.lock`. See
+[SECURITY.md](SECURITY.md) for vulnerability reporting and supply-chain
+practices.
+
+## Development
+
+```bash
+mypy tau/
+pyright tau/
+ruff check tau/
+ruff format tau/
+python -m pytest
+```
+
+See [Development Setup](docs/development.md) and
+[Contributing](CONTRIBUTING.md).
+
+## License
+
+Tau is licensed under the [MIT License](LICENSE).

tau_coding_agent-0.7.4/README.md

@@ -0,0 +1,247 @@
+<div align="center">
+
+  <img src="assets/wordmark.svg" alt="Tau" height="160">
+  <br>
+  <a href="https://pypi.org/project/tau-coding-agent/">
+    <img src="https://img.shields.io/pypi/v/tau-coding-agent.svg" alt="PyPI version">
+  </a>
+  <a href="https://pepy.tech/project/tau-coding-agent">
+    <img src="https://static.pepy.tech/badge/tau-coding-agent" alt="PyPI Downloads">
+  </a>
+  <a href="https://github.com/Jeomon/Tau/blob/main/LICENSE">
+    <img src="https://img.shields.io/badge/license-MIT-green" alt="License">
+  </a>
+  <img src="https://img.shields.io/badge/python-3.12%2B-blue" alt="Python">
+  <br>
+  <a href="https://github.com/Jeomon/Tau/actions/workflows/ci.yml">
+    <img src="https://github.com/Jeomon/Tau/actions/workflows/ci.yml/badge.svg" alt="CI status">
+  </a>
+  <a href="https://github.com/Jeomon/Tau/commits/main">
+    <img src="https://img.shields.io/github/last-commit/Jeomon/Tau.svg" alt="Last commit">
+  </a>
+
+</div>
+
+<br>
+
+Tau is a Python-based coding agent harness, heavily inspired by [Pi](https://github.com/earendil-works/pi) created by [Mario Zechner](https://github.com/badlogic). It combines an interactive terminal UI, multiple model providers, persistent sessions, tool execution, and an extension system in one package.
+
+> **Note:** There are several coding-agent projects also named
+> "Tau," including at least one that is itself a Python port of Pi. This
+> project (`tau`, [Jeomon/Tau](https://github.com/Jeomon/Tau)) was built
+> independently, taking inspiration only from the original
+> [Pi](https://github.com/earendil-works/pi) project. No other "Tau" project,
+> or any other Pi port, was referenced or used in its development.
+
+<p align="center">
+  <img src="assets/tui.jpeg" alt="Tau interactive terminal interface" width="700">
+</p>
+
+## Quick start
+
+Requires Python 3.12+.
+
+```bash
+pip install tau-coding-agent
+export NVIDIA_API_KEY=nvapi-...
+tau --provider nvidia
+```
+
+Then ask Tau to work in the current directory:
+
+```text
+Explain this repository, run its tests, and fix any failures.
+```
+
+**Other providers:** pass `--model <provider>/<model>` with the matching API
+key set, e.g. `GOOGLE_API_KEY=... tau --model google/gemini-2.5-flash`.
+
+## Commands
+
+### CLI usage
+
+```bash
+tau [OPTIONS] [MESSAGE]
+```
+
+```bash
+tau                                      # Start an interactive session
+tau --resume                             # Resume the latest session
+tau --resume abc123                      # Resume a specific session by ID
+tau --model claude-sonnet-4-6            # Start with a specific model
+tau --model groq/llama-3.3-70b-versatile # provider/model shorthand
+tau --print "Summarize this repository"  # Run once and print the result
+tau --mode json --prompt "Summarize this repo"  # Emit structured JSON events
+tau --mode rpc                           # Start JSON-RPC mode for IDE clients
+tau --ephemeral                          # Temporary session, nothing saved
+```
+
+Common flags:
+
+| Flag | Short | Description |
+|---|---|---|
+| `--prompt TEXT` | `-p` | Run a non-interactive prompt |
+| `--print` | | Print mode — run `MESSAGE` and exit (shorthand for `--mode print`) |
+| `--mode` | | `interactive` (default), `print`, `json`, `rpc` |
+| `--provider` | | Provider to use, e.g. `anthropic`, `openai`, `groq` |
+| `--model` | | Model ID, or `provider/model` shorthand |
+| `--resume [ID]` | `-r` | Resume the most recent or a specified session |
+| `--fork ID` | | Fork a specified session at startup |
+| `--ephemeral` | `-e` | Don't save this session to disk |
+| `--theme` | `-t` | UI theme: `dark`, `light`, or a custom theme |
+| `--cwd PATH` | `-c` | Set the working directory |
+| `--output-format` | `-f` | Non-interactive output: `text` or `json` |
+| `--quiet` | `-q` | Hide the non-interactive spinner |
+| `--version` | `-v` | Print the installed version |
+| `--help` | `-h` | Show help message |
+
+Full flag list, environment variables, and exit codes: [CLI reference](docs/cli-reference.md).
+
+### Subcommands
+
+```bash
+tau auth        # Manage provider credentials (login/logout, list)
+tau install      # Install a package (extension/skill/theme)
+tau remove        # Remove an installed package
+tau list           # List installed packages
+tau update          # Update installed packages
+```
+
+### Interactive slash commands
+
+Type these inside an interactive session (`tau`):
+
+| Command | What it does |
+|---|---|
+| `/new` | Start a fresh session |
+| `/resume` | Browse and resume a past session |
+| `/fork [entry-id]` | Branch the session tree at a specific entry |
+| `/tree` | Navigate the session tree, switch branches |
+| `/clone` | Duplicate the current session at the current position |
+| `/compact` | Summarize and compact the current context |
+| `/session` | Show session info, message counts, and stats |
+| `/model` | Pick a model by modality |
+| `/theme` | Open the theme picker |
+| `/effort` | Set the thinking effort level |
+| `/login` | Save credentials for a provider (API key or OAuth) |
+| `/logout` | Remove stored credentials for a provider |
+| `/clear` | Clear all messages from the current session |
+| `/copy` | Copy the last assistant message to the clipboard |
+| `/reload` | Reload extensions, skills, prompts, and settings |
+| `/settings` | Show current settings |
+| `/extensions` | Enable or disable extensions by scope |
+| `/watch <url> [question]` | Load public video metadata/captions via `yt-dlp` |
+| `/help` or `/?` | List all commands and keyboard shortcuts |
+| `/quit`, `/q`, or `/exit` | Exit Tau |
+
+Full interactive workflow guide: [Usage](docs/usage.md).
+
+## Compared to Pi
+
+| Area | Pi | Tau |
+|---|---|---|
+| Language | TypeScript | Python |
+| TUI rendering | Line-level diffing — rewrites a full line if any part of it changed | Cell-level diffing (`Buffer`/`Cell`, modeled after [ratatui](https://github.com/ratatui/ratatui)'s `Buffer::diff`) — only the changed cells within a row are redrawn |
+| LLM providers | ~40, including many CN/regional and gateway vendors | 14 major providers |
+| Audio (TTS/STT) | Not supported | ElevenLabs, Sarvam, Gemini, OpenAI |
+| Image/video generation | Not supported | OpenAI, Gemini, OpenRouter, Fal, Zai |
+| Sandboxing | microVM sandbox (Gondolin) is an example extension, excluded from the main build — the user wires it in themselves | `microsandbox` microVM ships as a builtin extension, enabled by default |
+| Packaging | 5 separately published npm packages | Single PyPI package |
+
+Core mechanics — built-in tools, session branching/compaction, extension and
+hook API, and the interactive/print/RPC execution modes — are functionally
+equivalent between the two.
+
+## Referencing files
+
+Type `@` in the interactive editor to search for a project file:
+
+```text
+Review @src/service.py and add tests for its error handling.
+```
+
+For one-shot execution, attach a file explicitly:
+
+```bash
+tau -p "Explain this file" @src/service.py
+```
+
+Tau also discovers project instructions from `AGENTS.md` and `CLAUDE.md`.
+See [Project Context Files](docs/project-context.md) for trust and discovery
+behavior.
+
+## Authentication and configuration
+
+Tau resolves provider credentials in this order:
+
+1. A programmatic runtime override
+2. A credential saved in `~/.tau/auth.json` (including keys saved by `/login`)
+3. A provider environment variable such as `ANTHROPIC_API_KEY`,
+   `OPENAI_API_KEY`, and `GOOGLE_API_KEY`
+
+Settings are merged in this order:
+
+1. Built-in defaults
+2. `~/.tau/settings.json`
+3. `.tau/settings.json`
+4. Environment variables
+5. Command-line options
+
+See [Authentication](docs/auth.md), [Installation](docs/installation.md), and
+[Inference Providers](docs/inference-providers.md) for provider-specific
+setup.
+
+## Documentation
+
+- [Quickstart](docs/quickstart.md) — First session in five minutes
+- [Usage](docs/usage.md) — Interactive workflows and commands
+- [CLI Reference](docs/cli-reference.md) — Command-line options and modes
+- [Inference Providers](docs/inference-providers.md) — Providers and speech timestamps
+- [Sessions](docs/sessions.md) — Persistence, branching, and compaction
+- [Tools](docs/tools.md) — Built-in and custom tools
+- [Extensions](docs/extensions.md) — Tools, commands, hooks, and plugins
+- [Terminal UI](docs/tui.md) — Rendering, Markdown, math, and components
+- [Python API](docs/python-api.md) — Embed Tau in another application
+- [Architecture](docs/architecture.md) — Internal design and data flow
+
+The complete documentation index is available at [docs/index.md](docs/index.md).
+
+## Install from source
+
+```bash
+git clone https://github.com/Jeomon/Tau.git
+cd Tau
+pip install -e .
+tau
+```
+
+## Security
+
+Tau executes enabled tools with the operating-system permissions of the process
+that launched it. The built-in `sandbox` extension routes terminal execution
+through a `microsandbox` microVM by default, but requires the `microsandbox`
+package and a supported platform — otherwise it falls back to unsandboxed host
+execution. Review project instructions and commands before approving work in
+untrusted repositories, and verify the sandbox is actually active (`/sandbox`)
+when stronger isolation matters.
+
+Dependency versions are pinned and recorded in `uv.lock`. See
+[SECURITY.md](SECURITY.md) for vulnerability reporting and supply-chain
+practices.
+
+## Development
+
+```bash
+mypy tau/
+pyright tau/
+ruff check tau/
+ruff format tau/
+python -m pytest
+```
+
+See [Development Setup](docs/development.md) and
+[Contributing](CONTRIBUTING.md).
+
+## License
+
+Tau is licensed under the [MIT License](LICENSE).

{tau_coding_agent-0.7.2 → tau_coding_agent-0.7.4}/docs/auth.md

@@ -144,7 +144,8 @@ The provider ID used as the key in `auth.json` and the `{PROVIDER}_API_KEY` env
 | Google | `google` | `GOOGLE_API_KEY` |
 | Mistral | `mistral` | `MISTRAL_API_KEY` |
 | Fireworks AI | `fireworks` | `FIREWORKS_API_KEY` |
-| Ollama | `ollama` | `OLLAMA_API_KEY` |
+| Hugging Face | `huggingface` | `HUGGINGFACE_API_KEY` |
+| Ollama | `ollama` | None required (local server, default `http://localhost:11434`) |
 
 ## Next Steps
 

{tau_coding_agent-0.7.2 → tau_coding_agent-0.7.4}/docs/development.md

@@ -240,7 +240,7 @@ Create `.vscode/settings.json`:
 Recommended: Use a virtual environment:
 
 ```bash
-python3.13 -m venv venv
+python3.12 -m venv venv
 source venv/bin/activate  # On Windows: venv\Scripts\activate
 
 pip install -e .
@@ -272,7 +272,7 @@ Check your Python version:
 python --version
 ```
 
-Ensure it's 3.13 or higher.
+Ensure it's 3.12 or higher.
 
 ### Missing Dependencies
 

{tau_coding_agent-0.7.2 → tau_coding_agent-0.7.4}/docs/http-proxy.md

@@ -5,7 +5,7 @@ Tau supports HTTP proxy configuration for all HTTP/HTTPS traffic via **settings.
 ## Use Cases
 
 ### Corporate and Enterprise Environments
-Many organizations require all HTTP/HTTPS traffic to route through corporate proxies for security monitoring, compliance, and access control. The `httpProxy` setting allows tau to work in these restricted environments without manual environment variable configuration.
+Many organizations require all HTTP/HTTPS traffic to route through corporate proxies for security monitoring, compliance, and access control. The `http_proxy` setting allows tau to work in these restricted environments without manual environment variable configuration.
 
 ### Network Restrictions
 Some networks block direct access to external LLM provider APIs (Anthropic, OpenAI, Google, Bedrock, etc.). Proxy support enables routing through approved servers that have whitelist access.
@@ -25,9 +25,9 @@ Configure HTTP proxy in `~/.tau/settings.json` or `.tau/settings.json`. **Note:*
 
 ```json
 {
-  "httpProxy": {
+  "http_proxy": {
     "url": "http://proxy.example.com:8080",
-    "noProxy": "localhost,127.0.0.1,*.internal.example.com",
+    "no_proxy": "localhost,127.0.0.1,*.internal.example.com",
     "headers": {
       "Proxy-Authorization": "Bearer token123",
       "X-Custom-Header": "value"
@@ -38,7 +38,7 @@ Configure HTTP proxy in `~/.tau/settings.json` or `.tau/settings.json`. **Note:*
 
 **Fields:**
 - `url` (required) — Proxy URL for both HTTP and HTTPS requests
-- `noProxy` (optional) — Comma-separated hosts to bypass the proxy
+- `no_proxy` (optional) — Comma-separated hosts to bypass the proxy
 - `headers` (optional) — Custom headers for proxy authentication
 
 Settings here override corresponding environment variables (HTTP_PROXY, HTTPS_PROXY, NO_PROXY).
@@ -49,7 +49,7 @@ Some corporate proxies require custom authentication headers:
 
 ```json
 {
-  "httpProxy": {
+  "http_proxy": {
     "url": "http://proxy.example.com:8080",
     "headers": {
       "Proxy-Authorization": "Basic dXNlcm5hbWU6cGFzc3dvcmQ=",
@@ -63,7 +63,7 @@ Alternatively, embed credentials directly in the proxy URL:
 
 ```json
 {
-  "httpProxy": {
+  "http_proxy": {
     "url": "http://username:password@proxy.example.com:8080"
   }
 }
@@ -370,7 +370,7 @@ OAuth login/refresh operations (e.g., GitHub Copilot, OpenAI, Anthropic) also re
 
 ### Notes
 
-- The `httpProxy` setting is **global-only** and cannot be overridden at the project level
+- The `http_proxy` setting can be set globally (`~/.tau/settings.json`) or per-project (`.tau/settings.json`); project settings are merged over global settings
 - Different providers may have additional proxy-related configuration options (e.g., Amazon Bedrock VPC endpoints)
 - Proxy validation occurs at startup — invalid proxy URLs will raise an error before any requests are made
 - SOCKS and PAC proxies are not supported; only HTTP/HTTPS proxies are allowed

{tau_coding_agent-0.7.2 → tau_coding_agent-0.7.4}/docs/inference-providers.md

@@ -8,11 +8,12 @@ Tau supports the following inference providers:
 
 Built-in API-key providers are `openai`, `anthropic`, `google`, `nvidia`,
 `groq`, `openrouter`, `perplexity`, `xai`, `bedrock`, `kimi`, `minimax`,
-`cerebras`, `deepseek`, `zai`, `kilocode`, `fireworks`, and `mistral`.
+`cerebras`, `deepseek`, `zai`, `kilocode`, `fireworks`, `huggingface`, and
+`mistral`.
 
 Tau also includes local `ollama`, Google/Anthropic/OpenAI-compatible Vertex AI
 providers, and OAuth providers for OpenAI Codex, Claude Code, GitHub Copilot,
-and Google Antigravity. Extensions can register additional providers.
+Google Antigravity, and xAI Grok. Extensions can register additional providers.
 
 The model picker is the authoritative model catalogue. Built-in model metadata
 changes more frequently than this guide, so model IDs and pricing are not
@@ -132,6 +133,34 @@ tau --model fireworks/accounts/fireworks/models/llama-v3p3-70b-instruct -p "Say
 - Uses the OpenAI-compatible completions API under the hood
 - Session affinity is maintained automatically via a per-request `x-session-affinity` header
 
+## Hugging Face
+
+Hugging Face's Inference Providers router gives OpenAI-compatible access to
+open-source models (DeepSeek, Llama, Qwen, Mixtral, and more) hosted across
+its partner inference providers.
+
+### Setup
+
+1. Create an account at [Hugging Face](https://huggingface.co)
+2. Generate an access token at [huggingface.co/settings/tokens](https://huggingface.co/settings/tokens) (a `read` token is sufficient)
+3. Set the environment variable:
+
+```bash
+export HUGGINGFACE_API_KEY=hf_...
+```
+
+### Verify
+
+```bash
+tau --model huggingface/deepseek-ai/DeepSeek-V3-0324 -p "Say hello"
+```
+
+### Notes
+
+- Uses the OpenAI-compatible completions API under the hood, routed through `https://router.huggingface.co/v1`
+- Model IDs are Hugging Face Hub repo IDs (e.g. `meta-llama/Llama-3.3-70B-Instruct`)
+- Built-in model entries are pinned to a specific backend with `<repo>:<provider>` (e.g. `openai/gpt-oss-120b:groq`) rather than the router's default `:fastest` routing, since unpinned routing can silently switch to a backend with different tool-calling behavior. You can still use any model/backend combination the router supports via `tau --model huggingface/<repo>[:<provider>]` even if it isn't in the built-in list.
+
 ## Ollama (Local)
 
 Run open-source models locally without API keys or internet.

{tau_coding_agent-0.7.2 → tau_coding_agent-0.7.4}/docs/keybindings.md

@@ -21,7 +21,7 @@
 | Shortcut | Action |
 |----------|--------|
 | Alt+Enter | Queue as follow-up message (delivered when agent is fully idle) |
-| Alt+Up | Restore queued messages into editor |
+| Ctrl+Up | Restore queued messages into editor |
 
 ### App
 
@@ -76,7 +76,7 @@ A `KeyMap` is `dict[str, list[str]]` — action name → list of key combos that
 | `tui.input.clear` | `ctrl+u` | Kill from cursor to start |
 | `tui.input.word_back` | `ctrl+w` | Delete previous word |
 | `app.message.followup` | `alt+enter` | Queue as follow-up message |
-| `app.message.dequeue` | `alt+up` | Restore queued messages into editor |
+| `app.message.dequeue` | `ctrl+up` | Restore queued messages into editor |
 | `app.details.toggle` | `ctrl+o` | Toggle thinking and tool-result previews |
 | `app.invocations.toggle` | `ctrl+e` | Toggle template and skill blocks |
 | `tui.app.quit` | `ctrl+c`, `ctrl+d` | Quit tau |