> ## Documentation Index
> Fetch the complete documentation index at: https://coding-1afcb9be.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Tools reference

> Tool inventory, profiles, and key input schemas from the runtime registry.

# Tools reference

This page lists the tools implemented by `coding_tools_mcp.server`. Tool availability depends on `--tool-profile`.

## Profiles

| Profile               | Exposed tools                                                                                   |
| --------------------- | ----------------------------------------------------------------------------------------------- |
| `full`                | All implemented tools.                                                                          |
| `read-only`           | Inspection tools only. Mutating process, patch, cwd, and permission tools are hidden.           |
| `compat-readonly-all` | Compatibility mode that marks exposed tools as read-only even when some can mutate local state. |

## Server and workspace tools

| Tool                     | Purpose                                                                                | Important inputs |
| ------------------------ | -------------------------------------------------------------------------------------- | ---------------- |
| `server_info`            | Returns server metadata, protocol information, active profile, and capability details. | none             |
| `check_exec_environment` | Reports command execution environment and confinement status.                          | none             |
| `get_default_cwd`        | Returns the runtime default working directory.                                         | none             |
| `set_default_cwd`        | Changes the default working directory for subsequent relative command operations.      | `path`           |

`set_default_cwd` is not exposed by the `read-only` profile.

## File and search tools

| Tool          | Purpose                                        | Important inputs                                                                                             |
| ------------- | ---------------------------------------------- | ------------------------------------------------------------------------------------------------------------ |
| `read_file`   | Reads UTF-8 text from a workspace file.        | `path`, `start_line`, `end_line`, `max_bytes`, `encoding`                                                    |
| `list_dir`    | Lists direct or recursive directory entries.   | `path`, `recursive`, `max_depth`, `max_entries`, `include_hidden`, `include_ignored`, `sort`                 |
| `list_files`  | Finds files by glob-style patterns.            | `path`, `patterns`, `glob`, `exclude_patterns`, `include_hidden`, `include_ignored`, `max_results`, `sort`   |
| `search_text` | Searches text using literal or regex matching. | `query`, `path`, `regex`, `case_sensitive`, `include_globs`, `exclude_globs`, `context_lines`, `max_results` |

Notes:

* `read_file` supports only `utf-8` and rejects binary-looking files.
* `read_file.max_bytes` is capped at 1 MiB.
* `search_text` uses `rg` when available and falls back to Python scanning.
* Default ignored names include `.git`, `node_modules`, build outputs, virtualenvs, and common cache directories.

## Editing tool

| Tool          | Purpose                                            | Important inputs   |
| ------------- | -------------------------------------------------- | ------------------ |
| `apply_patch` | Applies a unified patch to files in the workspace. | `patch`, `dry_run` |

`apply_patch` is hidden in `read-only`. Use `dry_run` to validate a patch without writing files.

## Command tools

| Tool           | Purpose                                    | Important inputs                                                                           |
| -------------- | ------------------------------------------ | ------------------------------------------------------------------------------------------ |
| `exec_command` | Starts or runs a command in the workspace. | `cmd`, `workdir`, `timeout_ms`, `yield_time_ms`, `max_output_bytes`, `stdin`, `tty`, `env` |
| `write_stdin`  | Sends input to a running command session.  | `session_id`, `chars`, `yield_time_ms`, `max_output_bytes`                                 |
| `kill_session` | Terminates a running command session.      | `session_id`, `signal`, `wait_ms`, `max_output_bytes`                                      |

Command behavior is shaped by permission mode. In `safe`, network-looking commands, shell expansion, inline scripts, sensitive environment values, and broad temp writes are blocked. In `trusted`, local development workflows are allowed while secret filtering and Landlock remain active when available. In `dangerous`, command permission gates are disabled.

## Git tools

| Tool         | Purpose                                     | Important inputs                                                     |
| ------------ | ------------------------------------------- | -------------------------------------------------------------------- |
| `git_status` | Returns repository status.                  | `path`, `include_untracked`, `max_entries`                           |
| `git_diff`   | Returns staged or unstaged diffs.           | `path`, `paths`, `staged`, `unstaged`, `context_lines`, `max_bytes`  |
| `git_log`    | Returns recent commits.                     | `path`, `ref`, `max_count`, `skip`                                   |
| `git_show`   | Shows a revision, optionally with diff.     | `rev`, `path`, `paths`, `include_diff`, `context_lines`, `max_bytes` |
| `git_blame`  | Returns blame information for a line range. | `path`, `rev`, `start_line`, `end_line`, `max_lines`                 |

These are Git inspection tools. They do not commit, checkout, reset, push, or mutate repository state.

## Permission tool

| Tool                  | Purpose                                                                                      | Important inputs                                                         |
| --------------------- | -------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------ |
| `request_permissions` | Requests permission for a gated operation such as command network access or patch execution. | `tool_name`, `permission`, `reason`, `arguments`, `scope`, `ttl_seconds` |

The accepted `tool_name` values are `exec_command` and `apply_patch`. Permission names include `network`, `destructive_command`, `long_timeout`, `sensitive_env`, `shell_expansion`, `inline_script`, `privileged_executable`, and `write_generated_or_ignored`.

## Image tool

| Tool         | Purpose                                                                    | Important inputs                                                        |
| ------------ | -------------------------------------------------------------------------- | ----------------------------------------------------------------------- |
| `view_image` | Reads a local image and returns either an MCP image payload or a data URL. | `path`, `max_bytes`, `max_width`, `max_height`, `auto_resize`, `output` |

`view_image` is enabled by default unless `CODING_TOOLS_MCP_ENABLE_VIEW_IMAGE=0` or `--enable-view-image` is not used in a wrapper that disables it. The input file must be within the workspace path boundary.

## Output convention

Tool calls return MCP content plus structured output. Successful tool payloads include `ok: true` where applicable. Failures are surfaced as structured errors with a code, message, category, retryability flag, and diagnostic details when available.