Environment Variables

Ripperdoc can be configured through environment variables. For model and credential settings, Ripperdoc only reads RIPPERDOC_* variables (provider-specific env vars like OPENAI_API_KEY are ignored). If you already set api_key in your config, you can skip RIPPERDOC_API_KEY.

RIPPERDOC_ Model Overrides

Any RIPPERDOC_* variable overrides the corresponding config field in memory for the current session (no changes are written to disk). If no profile exists, set at least one of RIPPERDOC_MODEL, RIPPERDOC_PROTOCOL, or RIPPERDOC_BASE_URL so Ripperdoc can construct a profile. When RIPPERDOC_PROTOCOL is not set, it is inferred from the base URL and/or model name.

Variable	Description
`RIPPERDOC_BASE_URL`	API base URL
`RIPPERDOC_API_KEY`	API key for the base URL
`RIPPERDOC_AUTH_TOKEN`	Bearer token (alternative to API key)
`RIPPERDOC_MODEL`	Model name (main pointer)
`RIPPERDOC_SMALL_FAST_MODEL`	Quick model name (quick pointer)
`RIPPERDOC_PROTOCOL`	Provider override (`anthropic`, `openai_compatible`, `gemini`)

Example:

RIPPERDOC_BASE_URL=https://api.example.com/v1
RIPPERDOC_API_KEY=YOUR_API_KEY
RIPPERDOC_MODEL=gpt-4o-mini
RIPPERDOC_PROTOCOL=openai_compatible

Behavior Settings

Timeouts and Retries

Variable	Description	Default
`RIPPERDOC_API_TIMEOUT`	API request timeout (seconds)	`120`
`RIPPERDOC_MAX_RETRIES`	Maximum API retry attempts	`10`
`RIPPERDOC_TOOL_TIMEOUT`	Per-tool execution timeout (seconds)	`300`
`RIPPERDOC_CONCURRENT_TOOL_TIMEOUT`	Total timeout for concurrent tool batches (seconds)	`600`

Context and Memory

Variable	Description	Default
`RIPPERDOC_CONTEXT_TOKENS`	Context window size limit	Model default
`RIPPERDOC_AUTO_COMPACT`	Enable automatic compaction	`true`
`RIPPERDOC_MAX_QUERY_ITERATIONS`	Maximum query loop iterations	`1024`
`RIPPERDOC_MESSAGE_WARNING_THRESHOLD`	Message count for warning	`500`
`RIPPERDOC_MESSAGE_CRITICAL_THRESHOLD`	Message count for critical warning	`1000`
`RIPPERDOC_RESUME_REPLAY_MAX_MESSAGES`	Max messages to replay on resume (-1 for all, 0 to skip)	`120`

File Operations

Variable	Description	Default
`RIPPERDOC_FILE_CACHE_MAX_ENTRIES`	Max cached file entries	`500`
`RIPPERDOC_FILE_CACHE_MAX_MEMORY_MB`	Max file cache memory (MB)	`50`
`RIPPERDOC_MAX_READ_FILE_SIZE_BYTES`	Max file size to read (bytes)	`262144`
`RIPPERDOC_MAX_READ_LINES`	Max lines to read without limit	`2000`

Bash Tool

Variable	Description	Default
`RIPPERDOC_BASH_TIMEOUT_MS`	Default bash timeout (ms)	`120000`
`RIPPERDOC_BASH_MAX_TIMEOUT_MS`	Maximum bash timeout (ms)	`600000`
`RIPPERDOC_BASH_MAX_OUTPUT_LENGTH`	Max output characters	`30000`
`RIPPERDOC_BASH_TASK_TTL_SEC`	Background task TTL (seconds)	`3600`

Shell Configuration

Variable	Description	Default
`RIPPERDOC_SHELL`	Preferred shell	Auto-detected
`RIPPERDOC_SHELL_PATH`	Path to shell executable	Auto-detected

LSP Settings

Variable	Description	Default
`RIPPERDOC_LSP_TIMEOUT`	LSP request timeout (seconds)	`20`

MCP Settings

Variable	Description	Default
`RIPPERDOC_MCP_MAX_OUTPUT_TOKENS`	Max MCP output tokens	`50000`
`RIPPERDOC_MCP_WARNING_TOKENS`	MCP warning threshold	`40000`

Agent Settings

Variable	Description	Default
`RIPPERDOC_AGENT_RUN_TTL_SEC`	Agent run time-to-live (seconds)	`3600`

Logging

Variable	Description	Default
`RIPPERDOC_LOG_LEVEL`	Log level (DEBUG, INFO, WARNING, ERROR)	`WARNING`

Development

Variable	Description	Default
`RIPPERDOC_DEBUG`	Enable debug mode	`false`
`RIPPERDOC_DEV_MODE`	Enable development mode	`false`

Example .env File

# Model overrides (env-only)
RIPPERDOC_BASE_URL=https://api.example.com/v1
RIPPERDOC_API_KEY=YOUR_API_KEY
RIPPERDOC_MODEL=gpt-4o-mini
RIPPERDOC_PROTOCOL=openai_compatible

# Timeouts
RIPPERDOC_API_TIMEOUT=180
RIPPERDOC_MAX_RETRIES=5

# Context
RIPPERDOC_CONTEXT_TOKENS=100000
RIPPERDOC_AUTO_COMPACT=true

# Bash settings
RIPPERDOC_BASH_TIMEOUT_MS=300000
RIPPERDOC_BASH_TASK_TTL_SEC=7200

# File settings
RIPPERDOC_MAX_READ_FILE_SIZE_BYTES=1048576
RIPPERDOC_MAX_READ_LINES=4000

# LSP settings
RIPPERDOC_LSP_TIMEOUT=30

# Logging
RIPPERDOC_LOG_LEVEL=WARNING

Loading Environment Variables

From Shell

export RIPPERDOC_BASE_URL="https://api.example.com/v1"
export RIPPERDOC_API_KEY="YOUR_API_KEY"
export RIPPERDOC_MODEL="gpt-4o-mini"
export RIPPERDOC_PROTOCOL="openai_compatible"
ripperdoc

From .env File

Create a .env file in your project root:

RIPPERDOC_BASE_URL=https://api.example.com/v1
RIPPERDOC_API_KEY=YOUR_API_KEY
RIPPERDOC_MODEL=gpt-4o-mini
RIPPERDOC_PROTOCOL=openai_compatible

Then use a tool like direnv or source it manually:

source .env
ripperdoc

In Python SDK

import os
os.environ["RIPPERDOC_BASE_URL"] = "https://api.example.com/v1"
os.environ["RIPPERDOC_API_KEY"] = "YOUR_API_KEY"
os.environ["RIPPERDOC_MODEL"] = "gpt-4o-mini"
os.environ["RIPPERDOC_PROTOCOL"] = "openai_compatible"

from ripperdoc.sdk import query

Security Notes

Never commit API keys to version control
Use .env.local for local overrides (add to .gitignore)
Rotate keys if accidentally exposed
Use environment-specific keys for different environments

Environment Variables

On this page