Reference

Configuration

Nothing is required. The first-run wizard writes what it needs; everything else has a sensible default.

Where§

• ~/lampson/.lampson/config.json — provider, model, keys (sealed), preferences. Written by /setup, the web ⚙ and the provider pill. Shared by the terminal and the web UI.
• ~/lampson/.env — optional overrides for the same settings, plus the knobs below. .env wins over config.json.
• LAMPSON_HOME — move ~/lampson somewhere else.

Provider§

LAMPSON_PROVIDER=deepseek        # openai | deepseek | kimi | groq | grok | openrouter | ollama | anthropic | minimax | glm | glm-coding
LAMPSON_API_KEY=sk-...           # or LAMPSON_API_KEY_DEEPSEEK=… one per provider
LAMPSON_MODEL=deepseek-chat
LAMPSON_BASE_URL=…               # any compatible endpoint (no trailing slash)
LAMPSON_WIRE=openai              # openai | anthropic, when the preset cannot guess

Behaviour§

LAMPSON_PERMISSION=ask           # ask | yolo | strict (the default mode)
LAMPSON_MAX_STEPS=40             # tool steps per turn
LAMPSON_BUDGET_TOKENS=400000     # cost budget per turn (cached tokens count 10%)
LAMPSON_COMPACT_AT=80000         # compact the context above this estimate
LAMPSON_MAX_TOKENS=8192          # output cap per response (auto-doubles on an empty answer)
LAMPSON_TIMEOUT=180              # seconds per model call
LAMPSON_EXPLORE_CAP=12           # read-only calls in a row before a nudge to act

Shell (Windows)§

Commands run through Git's bash by default. To use something else:

LAMPSON_SHELL=C:\Program Files\Git\bin\bash.exe
LAMPSON_SHELL_FLAG=-c

Unattended runs and approvals§

LAMPSON_TZ=-03:00                # for "daily 09:00" (default: the system's)
LAMPSON_PUBLIC_URL=https://…     # how this Lampson is reached from outside
LAMPSON_WEBHOOK_URL=https://…    # POST for each pending approval
LAMPSON_WEBHOOK_SECRET=…         # HMAC-SHA256 signature of the body
LAMPSON_WEB_TOKEN=…              # let a non-local script call the web API (Bearer)

See Scheduled tasks.

Lampson as an engine§

LAMPSON_AGENT=build                         # the default agent (terminal, run, API without "to")
LAMPSON_CHANNEL_TELEGRAM_AGENT=tienda       # which agent answers each channel: LAMPSON_CHANNEL_<CHANNEL>_AGENT
LAMPSON_TELEGRAM_TOKEN=…                    # Telegram channel: token from @BotFather
LAMPSON_TELEGRAM_WEBHOOK_SECRET=…           # the secret Telegram sends with every webhook
LAMPSON_CHATWOOT_URL=… · LAMPSON_CHATWOOT_TOKEN=… · LAMPSON_CHATWOOT_WEBHOOK_SECRET=…
LAMPSON_WEBHOOK_SECRET=…                    # signs the API callbacks (as well as approvals)

Plugins get their own settings with "config" (LAMPSON_X → PLUGIN_X). Agents live in .lampson/agents/ (project) and ~/lampson/agents/ (global). → Agents · HTTP API · Channels

Computer use and the decision judge§

LAMPSON_COMPUTER_USE=1                # turn computer use on (off by default)
LAMPSON_COMPUTER_GRANT_PROFILE=1      # work in your open browser, signed in as you
LAMPSON_COMPUTER_MAX_TABS=5           # tabs per turn
LAMPSON_COMPUTER_DRIVER=…             # cua-driver's path when it is not found by itself
LAMPSON_JUDGE=jev                     # jev | laya | off
TYPESAFE_API_KEY=…                    # Jev's key
LAMPSON_JUDGE_MODEL=…                 # with laya: the checkpoint folder

See Computer use and Decision judge.

Per project§

Things you may want to commit, so your team gets them: .lampson/agents/ (agents), .lampson/plugins/ (plugins), .lampson/mcp.json, .lampson/lsp.json, skills/, AGENTS.md. Lampson writes nothing else into your project.