编码代理运行时

AnkaLoop 是一个开箱即用的编码代理运行时,集成了持久上下文、多代理协作、技能、记忆、MCP、钩子和自动化能力,可运行在命令行、HTTP/WebSocket 服务器以及 Telegram 上。

  • by tao12345666333
  • Python
  • 本地部署
  • 55 星标
  • 更新于
在 GitHub 查看 tao12345666333/amcp

使用说明

A batteries-included coding-agent runtime that keeps working across terminals, servers, and Telegram.

Quick start · Why AnkaLoop · Configuration · Docs · Contributing


AnkaLoop is for developers who want a useful coding agent now—not another framework to assemble. It combines a capable tool loop, persistent context, multi-agent delegation, skills, memory, MCP, hooks, and automation in one Python package that you can run locally or self-host.

python -m pip install ankaloop
# or: uv tool install ankaloop
anka init
anka

Why AnkaLoop

  • Useful on the first run — read, search, edit, patch, and execute code with built-in tools.
  • Persistent by design — sessions, searchable history, project rules, and long-term memory survive beyond a single prompt.
  • One runtime, multiple surfaces — use the same agent from the CLI, an HTTP/WebSocket server, or Telegram.
  • Built for real tasks — context compaction, progress events, cancellation, bounded tool use, and provider retries make long-running work observable and controllable.
  • Multi-agent without plumbing — delegate focused work to built-in explorer, planner, and coder agents.
  • Open and extensible — add MCP servers, reusable skills, slash commands, hooks, and custom agent specifications without replacing the built-in experience.
  • Provider-flexible — connect to OpenAI, Anthropic, GMI, or any OpenAI-compatible endpoint.

Quick start

AnkaLoop requires Python 3.11+ and credentials for a supported model provider.

Install from PyPI

# Install the latest stable release (with pip)
python -m pip install ankaloop

# Or install with uv
uv tool install ankaloop

# Configure a provider, then start in the current project
anka init
anka

# Or run one task and exit
anka --once "summarize this repository and suggest the next test to run"

The package installs both anka (recommended) and ankaloop commands.

Run without installing

uvx ankaloop init
uvx ankaloop

Install optional Telegram support

python -m pip install "ankaloop[telegram]"
# or: uv tool install "ankaloop[telegram]"
anka telegram setup

What is included

Area Capabilities
Coding loop File reading and search, patching and writing, shell execution, planning, todos, and subagent tasks
Context Persistent sessions, AGENTS.md rules, progressive loading, and smart compaction
Memory Durable facts, episodic history, persona files, and full-text session search
Agents Built-in coder, explorer, planner, and focused_coder roles
Interfaces Rich terminal UI, FastAPI HTTP/WebSocket server, and Telegram bot
Research Built-in web search/fetch plus MCP integration over stdio and HTTP/SSE
Extensions Skills, slash commands, hooks, custom YAML agent specs, and an event bus
Automation Cron-compatible jobs for systemd, Kubernetes, and external schedulers
Observability OpenTelemetry traces (GenAI semantic conventions) for turns, model calls, tools, hooks, and subagents

How it fits together

 CLI ───────────────┐
 HTTP / WebSocket ──┼──▶ Agent runtime ──▶ Tools / MCP ──▶ Your project
 Telegram ──────────┘         │
                              ├── Sessions & memory
                              ├── Skills & project rules
                              └── Subagents

All interfaces use the same core runtime. Conversation state is persistent, while project rules, skills, and memory provide context across requests.

Usage

# Chat and one-shot tasks
anka
anka --once "create a hello.py file"
anka -t explorer --once "find all TODO comments"
anka --agent path/to/agent.yaml

# Sessions and agents
anka --session my-session
anka --list-sessions
anka --list-types
anka --clear

# MCP
anka mcp tools --server custom
anka mcp call --server custom --tool example_tool --args '{"query":"rust async"}'

# Server and remote client
anka serve
anka attach http://localhost:8080

# Telegram
anka telegram setup
anka telegram start

Run anka --help or anka <command> --help for the complete command reference.

Configuration

Run anka init for the interactive provider wizard. Configuration lives in ~/.config/ankaloop/config.toml. Set ANKA_CONFIG_DIR_NAME=amcp to keep the legacy ~/.config/amcp/ path from pre-rebrand deployments.

OpenAI-compatible endpoint

[chat]
active_provider = "primary"

[chat.providers.primary]
api_type = "openai"
base_url = "https://api.example.com/v1"
model = "provider/model-name"

Keep credentials out of version control and provide the key through the environment:

export OPENAI_API_KEY="your-api-key"
anka --once "explain the architecture of this repository"

You can define multiple [chat.providers.<name>] profiles. Telegram administrators can list them with /models and switch the active profile with /model use <name>.

Runtime and tool settings
[chat]
request_timeout_seconds = 120
max_retries = 2
retry_base_delay_seconds = 0.5
# Reasoning effort for models that support it:
# none | minimal | low | medium | high | xhigh | max (unset or "auto" = provider default).
# Can also be set per provider profile or through ANKA_REASONING_EFFORT.
reasoning_effort = "high"
# Keep model reasoning on a separate channel (default): stream APIs emit `reasoning`
# frames and responses carry `reasoning_content`. Set true to restore the legacy
# behavior of folding reasoning into the message content.
reasoning_as_content = false
# Replay each assistant message's reasoning_content in conversation history.
# Off by default: providers ignore a replayed reasoning trace, so sending it only
# inflates prompt tokens. Enable (usually per provider profile) only for providers
# that reject requests lacking it.
requires_reasoning_echo = false
tool_loop_limit = 300
bash_tool_limit = 100
default_max_lines = 400
mcp_tools_enabled = true
write_tool_enabled = true
edit_tool_enabled = true
sync_tool_settle_timeout_seconds = 2.0
default_agent = "coder"

[context]
progressive_tools = true
progressive_skills = true
response_ratio = 0.30
MCP servers
# HTTP/SSE transport
[servers.remote]
url = "https://example.com/mcp"

# stdio transport
[servers.local]
command = "npx"
args = ["-y", "@some/mcp-server"]

Configured tools are exposed as mcp__<server>__<tool>. Tool names are normalized for providers with strict function-name requirements.

Tracing (OpenTelemetry)
[tracing]
enabled = true
exporter = "otlp"                  # or "console"
endpoint = "http://localhost:4318"
capture_content = false            # opt in to record prompts / tool arguments

Install the exporter with pip install "ankaloop[tracing]". Standard OTEL_* environment variables override the TOML values. See the tracing guide.

See the quick-start guide, skills and commands guide, hooks guide, and tracing guide for more configuration examples.

Multi-agent runtime

Agent Mode Purpose
coder Primary General coding agent with write access and delegation
explorer Subagent Fast, read-only codebase exploration
planner Subagent Read-only analysis and implementation planning
focused_coder Subagent Bounded implementation of a specific change

Select an agent with anka -t <name>. Primary agents can use the task tool to delegate independent work to subagents.

Skills, commands, and hooks

  • Skills inject reusable instructions and resources only when relevant.
  • Slash commands turn repeatable prompts into /command shortcuts with arguments, shell output, and file references.
  • Hooks validate, modify, block, or audit tool calls before and after execution.

Project extensions live under .ankaloop/:

.ankaloop/
├── commands/       # TOML slash commands
├── hooks.toml      # pre/post tool hooks
├── memory/         # project knowledge
└── skills/         # reusable SKILL.md packages

Self-hosted server

anka serve                         # loopback only, http://localhost:8080
anka serve --host 0.0.0.0          # requires authentication
anka attach https://agent.example  # connect from another terminal

The server exposes session, prompt, streaming, cancellation, timeline, tool, and agent APIs. Visit /docs on a running server for its OpenAPI interface.

Non-loopback binds require a server API key. Configure [server.auth] or pass --api-key; deploy behind TLS or a trusted reverse proxy.

Docker
docker build -t ankaloop .

# Safe loopback-only default
docker run -it ankaloop serve

# Expose with authentication
docker run -p 8080:8080 \
  -e ANKA_HOST=0.0.0.0 \
  -e ANKA_API_KEY=replace-with-a-secret \
  ankaloop serve

For protocol details, see the client/server architecture and API compatibility notes.

Telegram

The optional Telegram integration supports direct messages, groups and topics, allowlists, pairing, streaming responses, bounded queues, cancellation, and session switching.

python -m pip install "ankaloop[telegram]"
anka telegram setup
anka telegram start

Shared commands include /new, /clear, /cancel, /session list, and /session switch <id>.

Install from source

git clone https://github.com/tao12345666333/ankaloop.git
cd ankaloop

uv sync --extra dev --extra telegram
source .venv/bin/activate

Run the local quality suite:

ruff format --check src tests
ruff check src tests
mypy src/ankaloop --ignore-missing-imports
python -m pytest -q -m "not llm"

Tests marked llm make live provider calls and require credentials. See CONTRIBUTING.md before opening a pull request.

Deployment

Project status

AnkaLoop is under active development. Feedback, bug reports, documentation improvements, and focused pull requests are welcome.

License

Licensed under the Apache License 2.0.