Opencode MCP

Findings (2)

• lowLow. Minor, or common in normal tools. Shown for information.
  Possible example credential in docs/tests: generic-api-key

  3 matches in this file

  tests/helpers.test.ts:516
• lowLow. Minor, or common in normal tools. Shown for information.
  No manifest declaring network access and environment variables (agent.json)

Network activity

Tools exposed (87)

• opencode_health

  Check server health and version

• opencode_config_get

  Get the current opencode configuration

• opencode_config_update

  Update the opencode configuration. Pass a partial config object with fields to update.

• opencode_config_providers

  List all configured providers and their default models

• opencode_project_list

  List all projects known to the opencode server

• opencode_project_init

  Initialize or open a project directory to host an independent OpenCode session. Use this to create new empty folders, or to explicitly open preexisting projects on the MCP host machine for parallel code generation workloads. This tool operates on the local filesystem; create remote server directorie

• opencode_project_current

  Get the current active project

• opencode_session_list

  List all sessions

• opencode_session_create

  Create a new session. Optionally provide a parentID to create a child session, and a title.

• opencode_session_get

  Get details of a specific session by ID

• opencode_session_delete

  Delete a session and all its data

• opencode_session_update

  Update session properties (e.g. title)

• opencode_session_children

  Get child sessions of a session

• opencode_session_status

  Get status for all sessions (running, idle, etc.)

• opencode_session_todo

  Get the todo list for a session

• opencode_session_init

  Analyze the app and create AGENTS.md for a session. NOTE: This is a long-running operation that may take 30-60+ seconds depending on project size.

• opencode_session_abort

  Abort a running session

• opencode_session_fork

  Fork an existing session, optionally at a specific message

• opencode_session_share

  Share a session publicly

• opencode_session_unshare

  Unshare a previously shared session

• opencode_session_diff

  Get the diff for a session, optionally for a specific message

• opencode_session_summarize

  Summarize a session using a specified model. NOTE: This is a long-running operation that may take 30-60+ seconds.

• opencode_session_revert

  Revert a message in a session

• opencode_session_unrevert

  Restore all reverted messages in a session

• opencode_permission_list

  List all pending permission requests across all sessions. When a session is blocked waiting for approval (e.g. to run a shell command or access a file outside the project), it appears here. Respond with `opencode_session_permission`.

• opencode_session_permission

  Respond to a permission request in a session. Use `opencode_permission_list` to see pending requests. Reply values: 'once' (approve this request only), 'always' (approve this + future matching requests for this session), 'reject' (deny the request).

• opencode_session_search

  Search sessions by keyword in title. Useful for finding a specific session among many.

• opencode_message_list

  List all messages in a session with formatted output showing roles and content

• opencode_message_get

  Get details of a specific message in a session

• opencode_message_send

  Send a prompt message to a session and wait for the AI response. Use parts to send text, and optionally specify a model.

• opencode_message_send_async

  Send a prompt asynchronously and return its messageId. Pass sessionId and messageId to opencode_wait to observe this exact turn.

• opencode_command_execute

  Execute a slash command in a session (e.g. /init, /undo, /redo)

• opencode_shell_execute

  Run a shell command through the opencode session

• opencode_find_text

  Search for text patterns in project files (regex supported). Returns file paths, line numbers, and matching lines.

• opencode_find_file

  Find files and directories by name (fuzzy match)

• opencode_find_symbol

  Find workspace symbols by name (functions, classes, variables, etc.)

• opencode_file_list

  List files and directories at a path

• opencode_file_read

  Read the content of a file

• opencode_file_status

  Get status for tracked files (VCS changes: modified, added, deleted, etc.)

• opencode_provider_list

  List all configured providers with their connection status. Returns a compact summary — use opencode_provider_models to see models for a specific provider.

• opencode_provider_models

  List available models for a specific provider. Call opencode_provider_list first to see provider IDs.

• opencode_provider_auth_methods

  Get available authentication methods for all providers

• opencode_provider_oauth_authorize

  Start OAuth authorization for a provider

• opencode_provider_oauth_callback

  Handle OAuth callback for a provider

• opencode_auth_set

  Set authentication credentials for a provider (e.g. API key). Credentials are stored globally and shared across all projects.

• opencode_path_get

  Get the current working path of the opencode server

• opencode_vcs_info

  Get VCS (version control) info for the current project (branch, remote, status)

• opencode_instance_dispose

  Dispose the current opencode instance (shuts it down). WARNING: This is destructive and will terminate the server.

• opencode_agent_list

  List all available agents with their names, descriptions, and modes (primary/subagent)

• opencode_command_list

  List all available commands (built-in and custom slash commands)

• opencode_lsp_status

  Get the status of LSP (Language Server Protocol) servers

• opencode_formatter_status

  Get the status of configured formatters

• opencode_mcp_status

  Get the status of all MCP servers configured in opencode

• opencode_mcp_add

  Add an MCP server dynamically to opencode

• opencode_tool_ids

  List all available tool IDs that the LLM can use (experimental)

• opencode_tool_list

  List tools with JSON schemas for a given provider and model (experimental)

• opencode_log

  Write a log entry to the opencode server

• opencode_setup

  Check OpenCode status, provider configuration, and optionally initialize a project directory. Use this as the first step when starting work — it tells you what is ready and what still needs configuration.

• opencode_ask

  Ask OpenCode a question in one step. Creates a new session, sends your prompt, and returns the AI response. This is the easiest way to interact with OpenCode.

• opencode_reply

  Send a follow-up message to an existing session. Use this to continue a conversation started with opencode_ask or opencode_session_create.

• opencode_conversation

  Get the full conversation history of a session, formatted for easy reading. Shows all messages with their roles and content.

• opencode_sessions_overview

  Get a quick overview of all sessions with their titles and status. Useful to find which session to continue working in.

• opencode_context

  Get full project context in one call: current project, path, VCS info, config, and available agents. Useful to understand the current state before starting work.

• opencode_wait

  Wait for a session or durable job. Returns completed, failed, input_required, or a resumable timeout. Cancelling this wait stops observation; use job_cancel or session_abort to stop OpenCode.

• opencode_review_changes

  Get a formatted summary of all file changes made in a session. Shows diffs in a readable format.

• opencode_provider_test

  Quick-test whether a provider is working. Creates a temporary session, sends a trivial prompt, checks the response, and cleans up. Great for debugging auth issues.

• opencode_run

  Send a task and wait for its correlated response. Returns a durable job ID; an observation timeout leaves the remote job running and can be resumed with opencode_wait.

• opencode_fire

  Dispatch a durable background task and return its job, session and message IDs immediately. Use opencode_check or opencode_wait to observe; job_cancel stops the task.

• opencode_check

  Get current task status, pending input, todos and file counts. Uses session summary metadata instead of fetching patches. Pass jobId for exact durable turn correlation; detailed includes response text.

• opencode_status

  Get a quick status dashboard: server health, provider count, session count, and VCS info. Lighter than opencode_setup — good for at-a-glance checks.

• opencode_tui_append_prompt

  Append text to the TUI's prompt input field

• opencode_tui_submit_prompt

  Submit the current prompt in the TUI (equivalent to pressing Enter)

• opencode_tui_clear_prompt

  Clear the current prompt text in the TUI

• opencode_tui_execute_command

  Execute a slash command through the TUI (e.g. '/init', '/undo')

• opencode_tui_show_toast

  Show a toast notification in the TUI

• opencode_tui_open_help

  Open the help dialog in the TUI

• opencode_tui_open_sessions

  Open the session selector in the TUI

• opencode_tui_open_models

  Open the model selector in the TUI

• opencode_tui_open_themes

  Open the theme selector in the TUI

• opencode_events_poll

  Poll project events from OpenCode, or explicitly select global scope. Collects up to maxEvents within the duration. Connection failures are reported with any partial events; stopping observation does not abort remote work.

• opencode_question_list

  List pending OpenCode questions, optionally filtered to a session.

• opencode_question_reply

  Answer an OpenCode question request. Supply one array of selected labels or free text per question.

• opencode_question_reject

  Reject a pending OpenCode question explicitly.

• opencode_job_get

  Observe a durable job by ID, including pending inputs and final results.

• opencode_job_cancel

  Explicitly abort the OpenCode session owned by a job. This also stops other work in that session.

• opencode_job_list

  List locally retained jobs in this OpenCode server and credential scope.

• opencode_job_input

  Respond to a job's pending questions or permissions. Omit responses to request MCP forms when supported, or receive manual response instructions. Never approves automatically.