Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
BCP_EVALNoSet to on, 1, true, or yes to enable the page eval tool in the MAIN world. Desktop client toggles this in the preparation page; development default is off.off
BCP_HOSTNoWebSocket bind address. Set to 0.0.0.0 to allow remote browsers; use a real token in that case.127.0.0.1
BCP_PORTNoExtension WebSocket port.8787
BCP_TOKENNoHandshake shared secret. Must match the token configured in the extension sidebar.bcp-dev-token
BCP_LLM_URLNoLLM gateway URL used for CAPTCHA verification.http://127.0.0.1:${BCP_LLM_PORT:-11435}
BCP_LLM_HOSTNoLLM gateway bind address. Changing from loopback exposes the service externally.127.0.0.1
BCP_LLM_PORTNoOpenAI-compatible LLM gateway port.11435
BCP_MCP_PORTNoMCP /mcp HTTP port. Always bound to loopback.8788
CONSOLE_HOSTNoConsole listening host.0.0.0.0
CONSOLE_PORTNoConsole listening port.8790
DATABASE_URLNoPrisma database URL. Relative paths resolve from the Prisma schema directory.file:../../../resource/db/console.db
BCP_LLM_MODELNoModel used by the local Ollama channel.qwen3.5:4b
BCP_FORCE_PLANNoSet to on to reject all mutating actions when there is no active plan.off
BCP_LLM_ENABLENoSet to 0 to completely skip the LLM gateway and local models.1
BCP_LOCAL_EVALNoSet to on, 1, true, or yes to enable local.eval steps and the local_eval tool, which runs JavaScript in the backend process. Desktop client toggles this in the preparation page; development default is off.off
IMAGE_BED_REPONoGitHub image bed repository. Restart Console after changing.
BCP_LLM_NUM_CTXNoContext window for the local Ollama channel.32768
IMAGE_BED_OWNERNoGitHub image bed owner. Restart Console after changing.
IMAGE_BED_TOKENNoGitHub image bed token. Restart Console after changing.
WECLAW_API_ADDRNoWeClaw sidecar HTTP address.127.0.0.1:18011
BCP_BROWSER_ROOTNoRoot directory for the managed browser profile and extension artifacts.~/.bcp/browser
BCP_INTERNAL_URLNoURL used by Console to reach the backend internal API.http://127.0.0.1:8789
BCP_LLM_PROVIDERNoLLM provider selection: auto, ollama, or lmstudio.auto
IMAGE_BED_BRANCHNoGitHub image bed branch. Restart Console after changing.
BCP_EVENT_WEBHOOKNoURL to which captcha_appeared, captcha_cleared, and browser_offline events are actively POSTed.
BCP_INTERNAL_PORTNoLoopback internal API port used by Console.8789
BCP_INTERNAL_TOKENNoLoopback internal API token used by Console.bcp-internal-dev-token
BCP_RPC_TIMEOUT_MSNoTimeout for a single browser action in milliseconds.30000
BCP_SCREENSHOT_DIRNoAdditional directory allowed for screenshot savePath. The system temporary directory is always allowed.
BCP_WECLAW_DISABLENoSet to 1 to skip the WeClaw sidecar.off
BCP_ALLOWED_DOMAINSNoComma-separated navigation allowlist. javascript:, data:, file:, and similar protocols are always rejected.
BCP_BLOCKED_DOMAINSNoComma-separated navigation blocklist. javascript:, data:, file:, and similar protocols are always rejected.
BCP_LLM_OLLAMA_PORTNoInternal Ollama adapter port.11437
BCP_LLM_LMSTUDIO_PORTNoInternal LM Studio adapter port.11436
IMAGE_BED_PATH_PREFIXNoGitHub image bed path prefix. Restart Console after changing.
BCP_LLM_CALL_RETENTION_DAYSNoNumber of days to retain LLM call logs. Cleanup runs daily at 04:00; 0 disables cleanup.30

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Server capabilities have not been inspected yet.

Tools

Functions exposed to the LLM to take actions

NameDescription

No tools

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources