Configuration

Kandev has three distinct configuration surfaces:

• backend startup configuration: defaults, config.yaml, then environment variables;
• persistent product settings edited in the web UI and stored in the database; and
• executor, agent, repository, and workflow profiles stored through their own Settings pages.

This page is the startup-configuration reference. Executor-specific fields are covered in Executors, and deployment examples are in Docker, Kubernetes, and Run as a service.

Quick path

1. Use the built-in profile defaults for a first run.
2. Add config.yaml only for stable operator-wide settings.
3. Use environment variables for deployment-specific overrides and secrets.
4. Use the web UI for persistent product settings, agents, executors, and workflows.

Load order and lifecycle

At backend startup, later sources override earlier ones:

1. embedded production-profile values and built-in defaults;
2. the first existing config.yaml; and
3. environment variables.

The public launcher has no --config option. It searches for exactly these files, in order:

1. config.yaml in the process working directory;
2. <KANDEV_HOME_DIR>/config.yaml, or ~/.kandev/config.yaml when KANDEV_HOME_DIR is unset; and
3. /etc/kandev/config.yaml.

Kandev uses only the first existing candidate. It does not merge files. A missing candidate is skipped, but an unreadable, malformed, or invalid first candidate stops startup and names that file. A home configuration file cannot set homeDir, because it cannot relocate the directory that selected it.

Configuration is read at process start; there is no file watcher. Restart Kandev after changing YAML or environment variables. The CLI, desktop shell, service manager, Docker, or Kubernetes may set environment variables on the backend, so those values can override a file unexpectedly. Use the process/service/container environment as the final source of truth.

Environment-variable naming

Viper maps a nested YAML key by replacing . with _, adding KANDEV_, and uppercasing. It does not split camelCase words. For example:

database.dbName       -> KANDEV_DATABASE_DBNAME
repoClone.basePath    -> KANDEV_REPOCLONE_BASEPATH

Some common camelCase keys have explicit compatibility aliases. Use the documented exact names below; snake_case spellings not listed here are not equivalent.

Complete backend reference

Root and server

When server.host is unset, server.hosts may provide a YAML list of bind addresses. The launcher derives its health targets and access URL from this resolved set. It probes an IPv4 wildcard through 127.0.0.1 and keeps localhost as the default browser/access URL, while it probes and accesses an IPv6 wildcard through [::1]. The backend still listens on every interface for a wildcard bind. The current local product path must not be treated as an authenticated multi-user perimeter. For remote access, bind to loopback and use a trusted authenticated tunnel/proxy, or isolate the network at the deployment layer.

Database

SQLite is the supported default and enables WAL mode. PostgreSQL deployments must provision the database, network policy, TLS trust, backups, and credentials before starting Kandev. Passing the password in an environment variable avoids putting it in YAML but still exposes it to processes/administrators allowed to inspect the environment; use your platform's secret injection controls.

database.path is an advanced SQLite file-path override. The Settings → System → Database and Backups pages use that exact file, its WAL files, and the sibling backups/ directory. Restore stages <configured-database-path>.new, quiesces scheduling and active workers, validates the checkpoint result, closes the SQLite pool, and uses rollback-capable quarantine replacement for the configured file and WAL sidecars. Restart Kandev immediately after a successful restore. When the override is empty, the default path remains <home>/data/kandev.db and the backup directory remains <home>/data/backups/. Kandev does not move snapshots from another directory automatically. The System restore endpoint is SQLite-only; use PostgreSQL recovery tools for PostgreSQL.

Default SQLite continuity

When database.path is empty, startup checks both <home>/data/kandev.db and the legacy default <home>/kandev.db before it opens a writable database. If only the legacy database exists and is valid, Kandev copies it to the current default with a validated SQLite snapshot. The legacy database and its -wal and -shm files remain available for recovery.

If the current default has no task history but the legacy default has task history, startup stops and names both paths. Kandev does not modify either database. Preserve both files, then select the intended database explicitly with database.path or KANDEV_DATABASE_PATH before restarting. If both defaults contain task history, Kandev keeps the current default and does not merge the databases. An explicit database path bypasses legacy discovery.

One backend owns a Kandev home at a time. When SQLite uses a custom path outside that home, the backend also owns that database path, so separate homes alone do not permit concurrent backends against one SQLite file. Use a separate home and database for an intentional second instance. Ownership is released when the backend exits.

Database-only snapshots also omit <home>/data/master.key, the AES-256 key used to decrypt stored secrets. Preserve that owner-only key with an independently secured home/data backup; restoring the database without its matching key leaves encrypted credentials unreadable. See Operations.

Event bus and NATS

An external NATS URL moves event traffic across the configured network and can embed credentials/TLS parameters. Protect it as a secret where applicable, require TLS for untrusted networks, and keep namespaces distinct when deployments share one NATS server. clusterId does not provide isolation in the current implementation.

Docker runtime

The Docker socket is effectively root-equivalent on many hosts. Do not publish it or assume docker.tlsVerify secures a TCP daemon; it currently does not. Configure TLS through a supported Docker endpoint/environment and validate it independently, or keep the daemon local. See Docker and Executors.

Core agent service

The launcher starts agentctl, performs a one-time nonce handshake, and supplies the resulting per-launch token internally. Do not persist or proxy its bootstrap/auth state. Agent command, model, environment, permission, and MCP configuration belongs in agent profiles rather than this section.

Setup and launch timing

tasks.preparationTimeout controls how long Kandev allows repository setup and executor-profile prepare scripts to run. The value uses Go duration syntax, such as 90s, 10m, or 1h.

Only positive durations are accepted. An unset, invalid, zero, or negative value uses the 10m default. Kandev reads this environment variable when the backend starts, so restart the backend after changing it. The setting applies to Local, Worktree, Docker, Sprites, and SSH launches.

Runtime launch phases use the configured preparation timeout plus a fixed five-minute allowance for runtime creation and agentctl readiness. With the default, each launch-phase limit is 15m. Preparation scripts use a separate context, so earlier work such as Sprite uploads does not reduce their full 10m preparation budget. The environment variable overrides YAML. This is a startup setting, not a database or Settings value.

Capacity and managed-process startup settings

These settings are read at startup. Their environment aliases remain compatible overrides, and a YAML value is not copied into a public process environment variable. Managed agentctl children receive their resolved subset through a private process contract.

Authentication, Office, Plugins, and feature flags

Do not infer security from auth.jwtSecret: setting it currently does not turn the local server into an authenticated public service. Office's JWT key has a narrower, active purpose. Store both active secrets and third-party API keys in your deployment secret manager; never commit them in config.yaml.

voice.openAIApiKey / KANDEV_VOICE_OPENAI_API_KEY was removed. Voice Mode is now the Voice Mode plugin, and its transcription key lives in that plugin's own settings. Kandev ignores the old key and no longer serves /api/v1/transcribe.

Trusted proxies for X-Forwarded-For

server.trustedProxies and its KANDEV_TRUSTED_PROXIES override list the reverse proxies whose forwarded client IP Kandev trusts. YAML uses an array; the environment form is a comma-separated list of IP addresses or CIDR ranges, for example 10.0.0.0/8,192.168.0.0/16. IPv6 addresses and CIDRs are accepted. When the TCP peer of a request is in the list, the client IP is read from X-Forwarded-For (then X-Real-IP); otherwise those headers are ignored and the TCP peer address is used. The resolved IP feeds the login session record (Settings > Account > Security) and the login rate-limiter key.

For one stable reverse proxy, list its immediate TCP peer as an exact IP:

server:
  trustedProxies:
    - 10.0.0.5

For a controlled proxy network whose address changes, use the narrowest CIDR that contains the proxy peers:

server:
  trustedProxies:
    - 10.0.0.0/28

Use the proxy peer address named in the warning. Do not list the browser's client network. A trusted CIDR lets every directly connected host in that range supply forwarded identity headers, so use an exact IP when the proxy address is stable and never trust a broad private-network range by default.

Default: unset, meaning no trusted proxies. Forwarded headers are ignored entirely and the recorded client IP is always the TCP peer. This is the secure default: gin would otherwise trust every proxy by default, which lets a directly reachable backend accept a spoofed client IP.

Security implication: only list proxies you control that always overwrite the forwarded headers. A backend reached directly by a caller whose address falls inside a listed range can have X-Forwarded-For spoofed by that caller, which also defeats the ClientIP-keyed login rate limiter (login attempts are limited per IP+email). Callers outside the listed ranges still fall back to their own peer address.

An entry that is neither a valid IP nor a valid CIDR is rejected at startup with a warning naming the bad value, and the whole variable is ignored (fail closed: no partial trust). A trailing or doubled comma is treated the same way. The backend never crashes on a bad value.

The same list gates X-Forwarded-Host, which a proxy sends so the port-scoped session cookie can be resolved from the browser's original host. A proxy that sends it while missing from the list logs ignoring X-Forwarded-Host from untrusted peer with the peer address and the forwarded host. That warning is logged once per distinct peer and host, not per request, so it names the misconfiguration without flooding the log. Add the peer it names to KANDEV_TRUSTED_PROXIES if it is your reverse proxy.

The value is read once at startup and must reach the backend process. Set the environment override in the environment of the process that launches kandev, for example a systemd drop-in for kandev.service (systemctl --user edit kandev.service) or the container environment. The supervisor manifest at ~/.kandev/supervisor/launch.json is generated by the launcher on every launch and is not an environment configuration source.

Logging

Every backend launch writes to <home>/logs/backend-logs.log and prints that resolved path at startup. The active file appends across same-day restarts and accepts at most 16 MiB. Before a new entry would exceed that limit, Kandev closes the file as backend-logs-YYYY-MM-DD-NNNNNN.log and opens a new active file.

Active and closed backend files use at most 256 MiB in total. Kandev removes the oldest closed segments when needed, so high-volume periods keep the newest evidence instead of stopping file logging. Three UTC days is the maximum file age, not a reserved allocation for each day. The segment size, total budget, and maximum age are fixed and are not configurable. Legacy backend-logs-YYYY-MM-DD.log files remain readable during upgrades and count toward the total budget. Files are owner-only (0600) on Unix.

Normal launches write info and above to the file and warn and above to stdout. --debug writes debug and above to the file while stdout remains warn and above. --verbose writes info and above to both. The format default becomes JSON when KUBERNETES_SERVICE_HOST is non-empty or KANDEV_ENV is exactly production/prod; otherwise it is text.

Debug output may contain repository paths, subprocess output, prompts, file content, and tool-call data. Treat it as sensitive.

Repository, worktree, and clone paths

Discovery roots bound automatic filesystem traversal, so scope them narrowly. They do not authorize explicitly selected repository paths: Add Local Repository validates and saves the exact accessible Git repository the user chooses without widening automatic scans. Worktrees and clones can contain credentials or generated files ignored by Git; review repository copy-file and setup/cleanup settings before remote execution. See Git operations.

Debug configuration

KANDEV_DEBUG_DEV_MODE=true selects the dev profile. make dev sets that selector and defaults the browser title to Dev Kandev. make start-debug enables pprof and debug logging without selecting the dev profile, and defaults the browser title to Debug Kandev. Debug mode is high risk. It enables local diagnostic surfaces and implies KANDEV_DEBUG_AGENT_MESSAGES=true and KANDEV_DEBUG_PPROF_ENABLED=true when not explicitly locked by the environment. ACP JSONL frames include complete prompts, file content, and tool calls. Do not enable it on a shared or network-exposed backend.

Minimal examples

For a local-only CLI server with an isolated home:

KANDEV_SERVER_HOST=127.0.0.1 \
KANDEV_HOME_DIR="$PWD/.kandev-local" \
kandev --headless

For a file-based deployment, override only what is needed:

homeDir: "/srv/kandev"

server:
  host: "127.0.0.1"
  port: 38429

logging:
  format: "json"

repositoryDiscovery:
  roots:
    - "/srv/repositories"

This example changes automatic discovery only. An explicitly selected repository may live outside /srv/repositories if the Kandev process can access and validate it.

A complete shape, including compatibility fields, is:

homeDir: ""

server:
  host: "0.0.0.0"
  port: 38429
  readTimeout: 30
  writeTimeout: 30
  trustedProxies: []
  webInternalUrl: ""
  webTitlePrefix: ""

database:
  driver: "sqlite"
  path: ""
  host: "localhost"
  port: 5432
  user: "kandev"
  password: ""
  dbName: "kandev"
  sslMode: "disable"
  maxConns: 25
  minConns: 5

nats:
  url: ""
  clusterId: "kandev-cluster" # compatibility-only today
  clientId: "kandev-client"
  maxReconnects: 10

events:
  namespace: ""

docker:
  enabled: true
  host: "unix:///var/run/docker.sock" # use the Windows named pipe on Windows
  apiVersion: ""
  tlsVerify: false                    # compatibility-only today
  defaultNetwork: "kandev-network"  # compatibility-only today
  volumeBasePath: "/var/lib/kandev/volumes" # compatibility-only today

agent:
  standaloneHost: "localhost"
  standalonePort: 39429

tasks:
  preparationTimeout: "10m"

credentials:
  file: ""

limits:
  ghMaxConcurrent: 8
  gitMaxConcurrent: 12
  lspMaxConnections: 8

messageQueue:
  maxPerSession: 10

agentctl:
  idleTimeout: "1h"
  idleReaperInterval: "1m"
  notificationQueueCapacity: 131072

planning:
  coalesceWindowMs: 300000

observability:
  otlpEndpoint: ""

launcher:
  webPort: 0
  healthTimeoutMs: 45000
  noBrowser: false

auth:
  jwtSecret: ""       # compatibility-only for the main HTTP product path
  tokenDuration: 3600 # compatibility-only for the main HTTP product path

logging:
  level: "info"
  format: "text"

repositoryDiscovery:
  roots: []                # automatic scan roots; explicit paths need not be included
  maxDepth: 5

worktree:
  enabled: true
  defaultBranch: "main"    # compatibility-only today
  cleanupOnRemove: true    # compatibility-only today
  fetchTimeoutSeconds: 60
  pullTimeoutSeconds: 60

repoClone:
  basePath: ""

debug:
  devMode: false
  pprofEnabled: false

office:
  jwtSigningKey: ""
  schedulerTickMs: 5000

Copying this entire file is unnecessary and can freeze old defaults in a deployment. Keep only deliberate overrides. On Windows, do not copy the Unix Docker host/path literals from this example.

Runtime feature toggles

Settings → System → Feature Toggles manages startup-time flags:

The KANDEV_FEATURES_* values have no canonical YAML keys. They are selected from the embedded profile and can be overridden by the environment or the database-backed toggle controls in this section. Use Settings → System → Feature Toggles for persistent product changes.

UI changes are persisted in the database and require a restart. An explicitly set environment value wins and locks the UI control. Otherwise a database override wins over the embedded profile/default. Resetting a toggle removes its database override.

For a risky release feature, keep the flag off in the shipped profiles, enable it only on a selected install through an admin override or explicit environment, restart, and test it there. Promote the prod profile default only after the feature is ready for everyone; keep the registry entry as a kill-switch until the rollout is complete, then remove the live flag and move its key and environment variable to the runtime registry's append-only retired identities. Plugins are part of the base product and are not a runtime toggle.

The source checkout's make dev activates the embedded development profile, which enables Office, debug surfaces, ACP logging, and a mock agent; authentication and Claude background prompt handoff remain opt-in. Installed run/desktop builds select the safe production profile unless the environment explicitly opts in. E2E mock variables and routes are test-only and must never be enabled on a public deployment.

Credentials and product settings

The Unread Messages preference in Settings > Preferences > Task Behavior controls the Slack-style New divider in session transcripts. It defaults off for each user, persists with user settings, and takes effect immediately. Enabling it also allows that user's active transcript view to advance the session read cursor.

Most integrations, executor profiles, agent profiles, MCP servers, repository settings, and UI preferences are persistent database records edited under Settings. They are not fields in config.yaml. Secret values use an encrypted secret store backed by <home>/data/master.key; filesystem permissions, database backups, and key backup are part of the security boundary.

For headless injection, Kandev can also read agent credentials from the process environment by their required name (for example ANTHROPIC_API_KEY) or the KANDEV_-prefixed form. KANDEV_CREDENTIALS_FILE adds a fallback JSON provider:

{
  "ANTHROPIC_API_KEY": "replace-at-deployment-time",
  "OPENAI_API_KEY": "replace-at-deployment-time"
}

The file is loaded lazily, expects a flat string-to-string object, and is cached; restart after changing it. A missing file behaves as no file credentials, while unreadable or invalid JSON produces credential-resolution errors. The database secret store and environment providers are consulted before this file. Restrict file permissions to the Kandev service account and never commit it.

Profile environment variables are eventually injected into agent subprocesses or remote executor environments. Anyone who can edit a profile, inspect a remote host/container, enable debug frame logs, or run commands as the Kandev account may be able to access them. Use least-privilege, task-scoped credentials and rotate them after exposure.

YAML secrets and file permissions

Stable startup secrets can be stored in config.yaml, including database.password, nats.url, auth.jwtSecret, office.jwtSigningKey, and observability.otlpEndpoint. Kandev keeps secret values in typed configuration, never writes those values to logs, and warns when a selected secret-bearing file is readable by group or other users on Unix. The warning does not block startup. Use owner-only mode 0600 and prefer the deployment's secret injection mechanism when one is available. Do not commit secret values to a repository.

Advanced operator tuning

These settings are startup-only unless stated otherwise. Their YAML keys are the canonical file form; the environment aliases remain compatibility overrides.

Changing concurrency values trades process pressure against throughput and requires a restart. Under Settings > Task Behavior > Message Queue, an admin can save an install-wide capacity and independently control manual and automatic merging. The merge switches apply live and persist across restarts. messageQueue.maxPerSession resolves as environment, YAML, saved setting, then default; a non-negative YAML value or any valid environment value locks only capacity. A negative environment value means unlimited, while malformed environment input falls through to the lower-precedence source. 0 means unlimited. Lowering the live limit does not prune existing rows; new admissions remain blocked until the pending count drops below the limit, while retries of already accepted work remain eligible. The default-on automatic switch affects only later admissions and never bypasses capacity or sweeps existing rows.

The current OTLP exporter strips an http:// or https:// prefix from the configured endpoint and always uses WithInsecure(). Treat this as implementation-bound cleartext transport: send it only to a trusted private collector over a protected network, not directly across an untrusted network. The service name is kandev-agentctl, and spans can include task/session/execution IDs plus raw agent-event JSON truncated to 8192 characters. That payload can contain prompts, files, and tool data. Use collector-side access controls and retention accordingly.

ACP debug-log controls

These apply only when KANDEV_DEBUG_AGENT_MESSAGES=true:

Retention values must be positive integers; invalid/non-positive values use defaults. Directories and files use owner-only 0700/0600 modes on Unix. Rotation, age pruning, and file-count pruning bound normal growth, but these files remain highly sensitive and can exist inside a Docker executor rather than on the host.

Configuration exclusions

The catalog records variables that are not operator startup settings. They have no public YAML key:

• Internal launcher and process wiring: KANDEV_INTERNAL_CONFIG_FILE, KANDEV_INTERNAL_CONFIG_HOME_FILE, KANDEV_INTERNAL_AGENTCTL_STARTUP_CONFIG, KANDEV_LAUNCHER_PARENT_PID, KANDEV_BACKEND_PID_FILE, KANDEV_DESKTOP_NATIVE_NOTIFICATIONS, KANDEV_TASK_ID, KANDEV_SESSION_ID, and KANDEV_WORKSPACE_ID.
• Generated or packaging values: KANDEV_DESKTOP_HEALTH_TOKEN, KANDEV_BUNDLE_DIR, and KANDEV_WEB_DIST_DIR.
• Test and profile selectors: KANDEV_E2E_MOCK, KANDEV_MOCK_AGENT, KANDEV_MOCK_GITHUB, KANDEV_MOCK_GITLAB, KANDEV_MOCK_JIRA, KANDEV_MOCK_LINEAR, and AGENTCTL_AUTO_APPROVE_PERMISSIONS.
• Runtime flags and diagnostics: KANDEV_FEATURES_OFFICE, KANDEV_FEATURES_AUTH, KANDEV_FEATURES_CLAUDE_BACKGROUND_PROMPT_HANDOFF, KANDEV_FEATURES_CLAUDE_MID_TURN_STEERING, KANDEV_DEBUG_AGENT_MESSAGES, KANDEV_DEBUG_ACP_MAX_FILES, KANDEV_DEBUG_ACP_RETENTION_HOURS, KANDEV_DEBUG_ACP_MAX_FILE_BYTES, KANDEV_MCP_LOG_FILE, and KANDEV_DEBUG_LOG_DIR.

These values are generated, injected by a supervisor or workspace, selected by the test/profile system, or intentionally limited to diagnostics. They are not part of the stable YAML contract.

Validation and troubleshooting

Startup validation currently enforces:

• server.port: 1-65535;
• database.driver: sqlite or postgres;
• PostgreSQL port, non-empty user/database name, and supported SSL mode;
• positive auth.tokenDuration;
• logging level/format; and
• positive repositoryDiscovery.maxDepth.

Other fields can pass configuration validation and still fail later, for example an unreachable NATS/PostgreSQL/Docker endpoint, unwritable log path, nonsensical timeout, or incompatible pool sizes. A field appearing in the schema does not prove its subsystem is available.

If a value appears ignored:

1. confirm the exact environment spelling, especially camelCase keys;
2. inspect the launcher/service/container environment for an overriding value;
3. confirm the selected config.yaml candidate is the first existing file in the working-directory, home, system order;
4. restart the backend; and
5. check whether the field is marked compatibility-only above.

Use kandev --verbose to surface startup errors. Do not use --debug merely to diagnose a YAML typo on an exposed machine; verbose logs are usually sufficient.

Variables used only to assemble/test the runtime, such as KANDEV_WEB_DIST_DIR, KANDEV_DESKTOP_RUNTIME_DIR, mock/E2E switches, supervisor socket/manifest values, and bootstrap nonces, are internal implementation contracts, not supported deployment configuration. KANDEV_MCP_LOG_FILE remains a debug-only component variable without a YAML key. KANDEV_BUNDLE_DIR is the narrow exception documented for installer/package integration in CLI; end users should still let the installer set it.