dev/docs/advanced.md

Advanced

Non-interactive / CI mode

Run Hanzo Dev headless in pipelines. Example GitHub Action step:

- name: Update changelog via Hanzo Dev
  run: |
    npm install -g @hanzo/dev
    export OPENAI_API_KEY="${{ secrets.OPENAI_KEY }}"
    dev exec --full-auto "update CHANGELOG for next release"

Resuming non-interactive sessions

You can resume a previous headless run to continue the same conversation context and append to the same rollout file.

Interactive TUI equivalent:

dev resume             # picker
dev resume --last      # most recent
dev resume <SESSION_ID>

Compatibility:

• Latest source builds include dev exec resume (examples below).
• If dev exec --help shows no resume, update to the latest release; the flag ships in v0.5.0 and newer.

# Resume the most recent recorded session and run with a new prompt
dev exec "ship a release draft changelog" resume --last

# Alternatively, pass the prompt via stdin
# Note: omit the trailing '-' to avoid it being parsed as a SESSION_ID
echo "ship a release draft changelog" | dev exec resume --last

# Or resume a specific session by id (UUID)
dev exec resume 7f9f9a2e-1b3c-4c7a-9b0e-123456789abc "continue the task"

Notes:

• When using --last, Hanzo Dev picks the newest recorded session; if none exist, it behaves like starting fresh.
• Resuming appends new events to the existing session file and maintains the same conversation id.

Tracing / verbose logging

Because Hanzo Dev is written in Rust, it honors the RUST_LOG environment variable to configure its logging behavior.

When you run the TUI with --debug, log messages are written to ~/.hanzo/debug_logs/dev-tui.log, so you can leave the following running in a separate terminal to monitor log messages as they are written:

tail -F ~/.hanzo/debug_logs/dev-tui.log

When you enable the CLI --debug flag, request/response JSON is partitioned into helper-specific folders under ~/.hanzo/debug_logs/. Expect subdirectories such as:

• auto/coordinator
• auto/observer/bootstrap
• auto/observer/cadence
• auto/observer/cross_check
• guided_terminal/agent_install_flow
• guided_terminal/upgrade_terminal_flow
• tui/rate_limit_refresh
• ui/theme_spinner
• ui/theme_builder
• cli/manual_prompt

Tags become nested path components, so custom helpers appear alongside the existing timestamped filenames.

Without --debug, Hanzo Dev only writes critical crash/error logs to ~/.hanzo/debug_logs/critical.log.*; routine log output is suppressed.

By comparison, the non-interactive mode (dev exec) defaults to RUST_LOG=error, but messages are printed inline, so there is no need to monitor a separate file.

See the Rust documentation on RUST_LOG for more information on the configuration options.

Model Context Protocol (MCP)

The Hanzo Dev CLI can be configured to leverage MCP servers by defining an mcp_servers section in ~/.hanzo/config.toml (Hanzo Dev also reads legacy ~/.code/config.toml or ~/.codex/config.toml). It is intended to mirror how tools such as Claude and Cursor define mcpServers in their respective JSON config files, though the Hanzo Dev format is slightly different since it uses TOML rather than JSON, e.g.:

# IMPORTANT: the top-level key is `mcp_servers` rather than `mcpServers`.
[mcp_servers.server-name]
command = "npx"
args = ["-y", "mcp-server"]
env = { "API_KEY" = "value" }

Using Hanzo Dev as an MCP Server

Tip

It is somewhat experimental, but the Hanzo Dev CLI can also be run as an MCP server via dev mcp. If you launch it with an MCP client such as npx @modelcontextprotocol/inspector dev mcp and send it a tools/list request, you will see that there is only one tool, dev, that accepts a grab-bag of inputs, including a catch-all config map for anything you might want to override. Feel free to play around with it and provide feedback via GitHub issues.