MCP (Model Context Protocol) server plugin for 1C:EDT, enabling AI assistants (Claude, GitHub Copilot, Cursor, etc.) to interact with EDT workspace. - MCP Protocol 2025-11-25 - Streamable HTTP transport with SSE support - Project Information - List workspace projects and configuration properties - Error Reporting - Get errors, warnings, problem summaries with filters - Check Descriptio
npx mdskills install DitriXNew/edt-mcp@DitriXNew? Sign in with GitHub to claim this listing.Comprehensive MCP bridge for 1C:EDT with 67 tools, strong CI/testing, excellent setup docs
Build & Unit Tests, E2E, and MCP Conformance all run on stock GitHub-hosted runners (cloud CI) — no docker image, no self-hosted runner. E2E and Conformance run against EDT 2026.1 (currently build 2026.1.2): the setup step installs a headless EDT of that version on the runner via
p2 director. E2E additionally imports the test fixtures into an empty workspace via the plugin's headless bootstrap (EDT_MCP_IMPORT_PROJECTS) and skips the live-infobase tools, so no 1C platform is needed. Each badge reflects its latest run.
MCP (Model Context Protocol) server plugin for 1C:EDT, enabling AI assistants (Claude, GitHub Copilot, Cursor, etc.) to interact with EDT workspace.
Tip:
Contributing / making changes? Read CLAUDE.md first — it's the code-conduct "minefield map": hard don'ts and the stop-and-think-twice zones for this codebase (BM transactions, the bilingual ru/en model, cascading rename, etc.). Detailed how-to lives in the skills under
.claude/skills/.
Important:
EDT version compatibility: Built and CI-validated against 1C:EDT 2026.1 (Ruby), currently build 2026.1.2.
https://ditrixnew.github.io/EDT-MCP/Close your EDT (!) and run:
rem Here "%VER_EDT% = 2025.2.3+30" just for example - please, set YOUR actual version !
set VER_EDT=2025.2.3+30
"\your\path\to\EDT\components\1c-edt-%VER_EDT%-x86_64\1cedt.exe" -nosplash ^
-application org.eclipse.equinox.p2.director ^
-repository https://ditrixnew.github.io/EDT-MCP/ ^
-installIU com.ditrix.edt.mcp.server.feature.feature.group ^
-profileProperties org.eclipse.update.reconcile=true
Once the installation has been completed successfully, we will see the following:
After that, EDT will automatically monitor the update site and install available updates when detected.
As well, we can also manually check via Help → About → Installation Details → Select MCP → Update
The get_form_screenshot and get_form_layout_snapshot tools need EDT to be launched with the following JVM flag:
-DnativeFormBufferedLayoutRender=true
Without it, both tools return blank output (gray PNG / empty elements list).
Why: EDT's NativeRenderService reads nativeFormBufferedLayoutRender once at class-load time. If it was unset at JVM startup, the singleton HippoLayoutService is constructed without its offscreen buffer handler, the C++ form renderer never writes captureable pixels back to Java, and the screenshot helper falls through to an SWT Control.print() of the native window — which on Windows produces a gray rectangle. Setting the flag at runtime via reflection does not help because the singleton has already been built.
How to add it (persistent, recommended):
Close EDT.
Open 1cedt.ini (next to 1cedt.exe, e.g. C:\Program Files\1C\1CE\components\1c-edt-2025.2.6+4-x86_64\1cedt.ini).
After the -vmargs line, add:
-DnativeFormBufferedLayoutRender=true
Start EDT.
How to add it (one-shot, no install changes):
"\1cedt.exe" -data "" -vmargs -DnativeFormBufferedLayoutRender=true
The same flag is also recommended for production EDT use — it enables the buffered native renderer that EDT itself benefits from.
If your screenshots still come back blank after adding the flag, verify with -vmargs actually appears before it in 1cedt.ini (Eclipse stops parsing -vmargs block once it hits a non--D line) and that EDT was fully restarted.
Go to Window → Preferences → MCP Server. The settings page has two tabs:
Note:
Display language. The settings page and the tag dialogs are bilingual (Russian / English) and follow the Eclipse/EDT display language — the same
-nllaunch argument (or OS locale) that localizes the rest of EDT. Launch EDT with-nl rufor Russian or-nl enfor English; any other locale falls back to English. The MCP tool surface itself (tool names, descriptions and errors) stays English regardless, as it is the AI wire contract.
Manage which tools are available to AI assistants. Tools are organized into groups that can be enabled or disabled together. See Tool Management for details.
The MCP server status bar shows real-time execution status with interactive controls.
Status Indicator:
User Signal Controls - Send signals to AI agent during tool execution
During Tool Execution:
MCP: update_database)When a tool is executing, you can send signals to the AI agent to interrupt the MCP call:
| Button | Description | When to Use |
|---|---|---|
| Cancel Operation | Stops the MCP call and notifies agent | When you want to cancel a long-running operation |
| Retry | Tells agent to retry the operation | When an EDT error occurred and you want to try again |
| Continue in Background | Notifies agent the operation is long-running | When you want agent to check status periodically |
| Ask Expert | Stops and asks agent to consult with you | When you need to provide guidance |
| Send Custom Message... | Send a custom message to agent | For any custom instruction |
How it works:
USER SIGNAL: Your message here
Signal Type: CANCEL
Tool: update_database
Elapsed: 20s
Note: The EDT operation may still be running in background.
Use cases:
Control which MCP tools are exposed to AI assistants. This lets you reduce context window usage and restrict AI to read-only operations when needed.
All 67 tools are organized into 9 semantic groups:
| Group | Description | Tools |
|---|---|---|
| Core / Project | EDT version, server self-diagnosis, on-demand tool guides, project listing, configuration, validation, XML export/import, project removal, project creation (configuration / extension / externalObjects) | get_edt_version, get_server_status, get_tool_guide, list_projects, get_configuration_properties, clean_project, revalidate_objects, get_check_description, export_configuration_to_xml, import_configuration_from_xml, delete_project, create_project |
| Errors & Problems | Error reporting and workspace markers (bookmarks, tasks) | get_problem_summary, get_project_errors, get_markers |
| Code Intelligence | Content assist, documentation, metadata browsing | get_content_assist, get_platform_documentation, get_metadata_objects, get_metadata_details, list_subsystems, get_subsystem_content, find_references |
| Tags | Tag management | get_tags, get_objects_by_tags |
| Applications & Testing | App management, infobase create/delete, database updates, launch, termination, testing | get_applications, create_infobase, delete_infobase, list_configurations, update_database, debug_launch, terminate_launch, run_yaxunit_tests |
| Debugging | Breakpoints, stepping, variable inspection and modification | set_breakpoint, remove_breakpoint, list_breakpoints, wait_for_break, get_variables, set_variable, step, resume, evaluate_expression, debug_yaxunit_tests, debug_status, start_profiling, stop_profiling, get_profiling_results |
| BSL Code | Module browsing, code reading/writing, search, form layout inspection | read_module_source, write_module_source, get_module_structure, list_modules, search_in_code, read_method_source, get_method_call_hierarchy, go_to_definition, get_symbol_info, get_form_layout_snapshot, get_form_screenshot, get_template_screenshot, validate_query |
| Refactoring | Metadata create (objects, members and form members), rename, delete, set properties, adopt into an extension | create_metadata, rename_metadata_object, delete_metadata, modify_metadata, adopt_metadata_object |
| Translation (LanguageTool) | Translation strings generation, configuration synchronization, project info | generate_translation_strings, translate_configuration, get_translation_project_info |
Enable or disable entire groups or individual tools from the Tools tab in Window → Preferences → MCP Server. Disabled tools are filtered out of tools/list responses. If a client calls a disabled tool directly through tools/call, the server returns a message explaining that the tool is disabled.
Quickly switch between common tool configurations using presets:
| Preset | Description |
|---|---|
| All Tools | All 67 tools enabled (default) |
| Analysis Only | Read-only analysis — Core, Errors, Code Intelligence, Tags |
| Code Review | Analysis + BSL code reading (excludes write_module_source) |
| Development | Full development without debugging tools |
Select a preset from the dropdown in the Tools tab. The preset auto-detects based on the current enabled/disabled state and shows "Custom" when the configuration doesn't match any built-in preset.
Some tools have configurable default values for parameters like result limits. These defaults are used when the AI client doesn't specify the parameter explicitly:
| Tool | Parameter | Default | Range |
|---|---|---|---|
get_project_errors | Result limit | 100 | 1–1000 |
get_markers | Result limit | 100 | 1–1000 |
get_metadata_objects | Result limit | 100 | 1–1000 |
get_content_assist | Result limit | 100 | 1–1000 |
search_in_code | Max results | 100 | 1–500 |
search_in_code | Context lines | 2 | 0–5 |
read_module_source | Max lines | 500 | 100–50000 |
terminate_launch | Termination timeout (sec) | 10 | 1–120 |
Configure these in the Tools tab by selecting a tool that has configurable parameters — the parameter editors appear in the details panel below the tool tree.
For clients that work better with a small initial tool surface, the server can expose
only a core toolset up front and reveal the rest on demand — shrinking the
always-loaded tools/list. This is off by default (the full list is exposed,
unchanged); turn it on with the Progressive disclosure preference (or the
EDT_MCP_PROGRESSIVE_DISCLOSURE=true environment variable for headless/CI runs).
When on, only the core toolset (navigation, source read, metadata discovery, and the
two management tools) appears in tools/list. To use more tools:
list_toolsets to see the groups (core, metadata, code, debug,
testing, profiling, forms, tags, translation, project) and their tools.enable_toolset with toolsets=[ids] (e.g. ["code","debug"]).tools/list — the revealed tools now appear.This server does not push notifications/tools/list_changed, so the client must
re-list after enabling (the enable_toolset response says so). A hidden tool is only
hidden from tools/list; it remains callable by name. These progressive-disclosure
toolsets are distinct from the Tool Groups above (which the Tools tab uses to
enable/disable tools persistently).
Create .vscode/mcp.json:
{
"servers": {
"EDT MCP Server": {
"type": "sse",
"url": "http://localhost:8765/mcp"
}
}
}
Other AI Assistants - Cursor, Claude Code, Claude Desktop
Note: Cursor doesn't support MCP embedded resources. Enable "Plain text mode (Cursor compatibility)" in EDT preferences: Window → Preferences → MCP Server.
Create .cursor/mcp.json:
{
"mcpServers": {
"EDT MCP Server": {
"url": "http://localhost:8765/mcp"
}
}
}
Note: By editing the file
.claude.jsoncan be added to the MCP either to a specific project or to any project (at the root). If there is no mcpServers section, add it.
Add to .claude.json (in Windows %USERPROFILE%\.claude.json):
"mcpServers": {
"EDT MCP Server": {
"type": "http",
"url": "http://localhost:8765/mcp"
}
}
claude_desktop_config.json accepts only stdio servers (a command), not a direct url, so bridge the HTTP endpoint with mcp-remote (requires Node.js):
{
"mcpServers": {
"EDT MCP Server": {
"command": "npx",
"args": ["-y", "mcp-remote", "http://localhost:8765/mcp"]
}
}
}
The Settings → Connectors ("custom connector") route needs an HTTPS server reachable from Anthropic's cloud, so a plain-HTTP
localhostendpoint must go through themcp-remotestdio bridge above. You could expose it over an HTTPS tunnel (ngrok / Cloudflare Tunnel) to use a connector — but this server can write code into your configuration and start a debugger, so putting it on the public internet is a serious security risk and is not recommended.
{
"mcpServers": {
"EDTMCPServer": {
"type": "streamableHttp",
"url": "http://localhost:8765/mcp"
}
}
}
{
"mcpServers": {
"EDTMCPServer": {
"serverUrl": "http://localhost:8765/mcp"
}
}
}
Running more than one EDT instance at once? edt-mcp-proxy is a standalone router that exposes a single, stable MCP endpoint on :8764 and forwards each call to the right EDT-MCP instance by projectName, discovering live instances in the background. It ships as edt-mcp-proxy-.jar alongside the plugin archive in every release. See proxy/README.md for setup, CLI options and configuration.
The full per-tool reference lives in docs/tools/ — one page per
tool (what it does, every parameter, and how it works), generated from the live MCP
server (get_tool_guide) so it never drifts from the code. Index below; re-generate
with python docs/generate_tool_docs.py.
85 tools, grouped by toolset. Full per-tool pages under docs/tools/.
Always-on essentials: project/module navigation, source read, metadata discovery, and the toolset-management tools (list_toolsets / enable_toolset).
| Tool | Description |
|---|---|
enable_toolset | Reveal (or hide) tool groups for progressive disclosure. Pass toolsets=[ids] from list_toolsets to reveal them, then RE-REQUEST tools/list to see the newly r… |
get_edt_version | Returns the running 1C:EDT version as a plain version string. Returns "Unknown" when the version cannot be determined. |
get_metadata_details | Get detailed properties of one or more 1C metadata objects (basic info by default, or every reflected section with 'full: true'). Use it after get_metadata_o… |
get_metadata_objects | Get a flat list of 1C configuration metadata objects (Name, Synonym, Comment, Type, ObjectModule, ManagerModule) as a Markdown table. Use it to discover what… |
get_module_structure | Get structure of a BSL module: all procedures/functions with signatures, line numbers, regions, execution context (&AtServer, &AtClient), export flag, and pa… |
get_server_status | Self-diagnosis snapshot of the running MCP server: listening port, MCP protocol version, plugin version, EDT version, enabled/total tool counts, the plainTex… |
get_tool_guide | Get the full on-demand how-to for a tool: its description, every parameter (type, required, allowed values) and extended examples/preconditions kept OUT of t… |
list_modules | List BSL modules in an EDT project as a table (module path, module type, parent type, parent name). Use it to discover module paths before reading or editing… |
list_projects | List all workspace projects with properties (name, path, type, natures). format='md' (default) returns the human Markdown table; format='json' returns the ma… |
list_toolsets | List the tool groups (toolsets) used by progressive tool disclosure: each toolset's id, title, description, member tools, and whether it is currently visible… |
read_module_source | Read BSL module source code from an EDT project, whole file or a line range. Returns YAML frontmatter (including a contentHash revision token to round-trip i… |
search_in_code | Literal/regex full-text search across all BSL modules in a project. Matching is purely textual and NOT ru/en dialect-aware, so a query in one BSL language wo… |
Metadata objects: discovery, create/modify/delete/rename/adopt, subsystems, configuration.
| Tool | Description |
|---|---|
adopt_metadata_object | Adopt a base-configuration metadata object or member (object / form / attribute / tabular section / ...) into a configuration EXTENSION so the extension can… |
create_launch_config | Create a 1C:EDT runtime-client launch configuration (thin/thick/web). The SAME config works for both run and debug (mode is chosen at launch time by debug_la… |
create_metadata | Create a metadata node addressed by a 1C full-name FQN: a top-level object (Catalog.Products) or a subordinate member (Catalog.Products.Attribute.Weight, Inf… |
delete_launch_config | Delete a 1C:EDT launch configuration by name (runtime client or Attach). Destructive: guarded by a confirm-preview - call without confirm to preview (no chan… |
delete_metadata | Delete a metadata node (object or member, including a FORM object 'Type.Object.Form.Name', a FORM member - item / attribute / command / handler - an XDTO pac… |
export_common_picture | Export a 1C CommonPicture (общая картинка) as PNG and list its picture variants (dpi, theme, interface variant, direction, template flag, glyph size). Resolv… |
get_configuration_properties | Get 1C:Enterprise configuration properties (name, synonym, comment, script variant, compatibility mode, etc.) |
get_subsystem_content | Get one 1C subsystem's content: properties, its metadata objects (Type/Name/Synonym/FQN) and child subsystems, identified by FQN (e.g. 'Subsystem.Sales.Subsy… |
list_common_pictures | List a 1C configuration's CommonPicture objects and the variants each carries in its Picture.zip (DPI, theme, interface variant, template flag, glyph size, p… |
list_configurations | List EDT launch configurations (runtime client + Attach + other 1C types) with their running state. This is the discovery step before debug_launch / run_yaxu… |
list_subsystems | List 1C subsystems of a configuration as a flat table (FQN, Synonym, Comment, InCommandInterface, content count, children count). Walks the whole tree by def… |
modify_metadata | Set properties of a metadata node (object or member, including a FORM member - item / attribute / command) addressed by a 1C full-name FQN, as properties=[{n… |
rename_metadata_object | Rename a metadata object or attribute, cascading the change across all references in BSL code, forms, and other metadata. Use the two-phase workflow: call wi… |
BSL code: write/read methods, call hierarchy, go-to-definition, references, content assist, queries.
| Tool | Description |
|---|---|
find_references | Find every place a metadata object is used: BSL code modules (with line numbers), other metadata, forms, roles, subsystems, etc. Pass the object FQN; the typ… |
get_content_assist | Get code-completion proposals at a 1-based line/column in a BSL module - the members, globals and variables valid at that caret (e.g. after a '.'). May retur… |
get_method_call_hierarchy | Find a BSL method's call hierarchy: who calls it (callers, default) or what it calls (callees), via semantic AST analysis that resolves ru/en spellings (unli… |
get_outgoing_structures | For each outgoing qualified call in a BSL module (or one method), report the top-level literal keys of the Structure passed as its first argument (local .Ins… |
get_symbol_info | Get type/hover info about a symbol at a position in a BSL module. Returns inferred types, signatures, and documentation. |
go_to_definition | Go to the definition of a symbol (the inverse of find_references): a qualified method 'ModuleName.MethodName', a bare 'MethodName' (also pass modulePath), or… |
read_method_source | Read a specific procedure/function from a BSL module by name. Returns source code with metadata. Lists available methods if not found. Use this for one metho… |
validate_query | Validate 1C:Enterprise query language (QL) text against a project, returning syntax and semantic errors with line numbers. Use to check a query before embedd… |
write_module_source | Write BSL source code to a 1C metadata object module. Use to edit a module: searchReplace a fragment (default, needs oldSource), replace the whole file, or a… |
Runtime debugging: launch/attach, breakpoints, step/resume, variables, expression evaluation.
| Tool | Description |
|---|---|
debug_launch | Start an EDT debug session: either an existing config by launchConfigurationName (runtime client OR Attach, the latter needed to debug server-side code), or… |
debug_status | Report active debug sessions: applicationId (real or synthetic 'attach:' / 'launch:'), launch configuration name/type, mode (debug/run), whether… |
evaluate_expression | Evaluate a BSL expression in the context of a suspended stack frame. Pass frameRef from wait_for_break and the expression text. WARNING: this executes arbitr… |
get_applications | Get list of applications (infobases) for a project. Returns application ID, name, type, and update state. Application ID is required for update_database and… |
get_variables | Read variables from a stack frame of a suspended debug thread. Pass frameRef from wait_for_break (preferred) or threadId+frameIndex. Use expandPath to drill… |
list_breakpoints | List active line breakpoints. Optionally filter by projectName. |
remove_breakpoint | Remove a 1C BSL line breakpoint. Either pass breakpointId (returned from set_breakpoint) or projectName+module+lineNumber to look it up by coordinates. |
resume | Resume a suspended debug thread or all threads of a debug target. Pass threadId (from wait_for_break) or applicationId. applicationId accepts ANY id form for… |
set_breakpoint | Set a line breakpoint on a 1C BSL module. Accepts either an EDT module-relative path (e.g. 'CommonModules/Foo/Module.bsl') or an absolute filesystem path. Us… |
set_variable | Set a BSL variable's value in a suspended debug frame. WRITE/side-effect: EXECUTES the entered value as a BSL literal/expression live in the running 1C appli… |
step | Step a suspended debug thread. kind ∈ {over, into, out}. Blocks until the next SUSPEND event (or timeout) and returns the new frame snapshot. |
terminate_launch | Terminate one or more 1C launches started from THIS EDT instance; externally launched 1C clients are never touched. Select ONE target mode: launchConfigurati… |
wait_for_break | Wait for a debug suspend event (e.g. breakpoint hit) on the given application. Returns the suspended thread/frame snapshot, or {hit:false} on timeout. applic… |
YAXUnit unit testing: run and debug test suites.
| Tool | Description |
|---|---|
debug_yaxunit_tests | Deprecated alias for run_yaxunit_tests with debug=true. Launches YAXUnit tests in DEBUG mode so breakpoints fire, then call wait_for_break to inspect. Prefer… |
run_yaxunit_tests | Run YAXUnit tests for a 1C:Enterprise project and return a JUnit Markdown report. Polls for up to timeout seconds, then returns the report or Pending (… |
Performance profiling: start/stop a measurement and read the results.
| Tool | Description |
|---|---|
get_profiling_results | Get profiling (performance measurement) results after a debug session: per-module, per-line call count, timing and percentage. Returns only the MOST RECENT m… |
start_profiling | Start performance measurement on the active debug target. Enables line-level profiling: call counts and timing for every executed BSL line. Start-only and id… |
stop_profiling | Stop performance measurement on the active debug target. Counterpart to start_profiling: deterministically switches profiling off. Idempotent: if profiling i… |
Form and template rendering: form layout snapshot, form screenshot, template screenshot.
| Tool | Description |
|---|---|
get_form_layout_snapshot | Return a YAML snapshot of a form's calculated WYSIWYG layout (bounds, element types, display properties) as text; use it to inspect or compare what a form ac… |
get_form_screenshot | Capture a PNG screenshot of a form's WYSIWYG editor; pass formPath to open the form automatically or omit it to shoot the active editor. Requires EDT launche… |
get_template_screenshot | Capture a PNG screenshot of a 1C template (a SpreadsheetDocument print form) as EDT renders it, so its layout and text are visible to an AI. Works for a comm… |
Tag-based organization: list tags and find objects by tag.
| Tool | Description |
|---|---|
get_objects_by_tags | Get metadata objects filtered by tags. Returns objects that have any of the specified tags, including tag descriptions and object FQNs. |
get_tags | Get list of all tags defined in the project. Tags are user-defined labels for organizing metadata objects. Returns tag name, color, description, and number o… |
Configuration translation via LanguageTool: extract, translate, project info.
| Tool | Description |
|---|---|
generate_translation_strings | Generate translation strings (.lstr/.trans/.dict) for a configuration project: scans translatable features and writes the resulting keys into the project's s… |
get_translation_project_info | Return LanguageTool metadata for a project: the translation storages declared on it and the available translation provider IDs. Use it to check whether a dic… |
translate_configuration | Run EDT 'Translate configuration' on a configuration project - reads the dictionaries from the storages bound to it (external dictionary storage projects wit… |
Project operations: clean/revalidate, update DB, export/import XML, problems and markers, docs.
| Tool | Description |
|---|---|
build_external_objects | Build (compile to disk) the external data processors/reports of an EDT external-object project to .epf/.erf files. Build ONE object with objectName, or ALL o… |
clean_project | Clean EDT project and trigger full revalidation. Direction: DISK -> MODEL - re-imports the on-disk src/ .mdo files into the in-memory model. Refreshes files… |
create_git_branch | Create a new local git branch, optionally check it out, and optionally attach an EXISTING infobase (application, from get_applications) to the new branch's c… |
create_infobase | Create a new FILE infobase (1C database) OR register an existing one, and bind it to a configuration project so it appears in get_applications. mode='create'… |
create_project | Create a NEW 1C project in the EDT workspace. projectKind selects the kind: 'configuration' (standalone), 'extension' (bound to a base configuration), or 'ex… |
delete_infobase | Remove a FILE infobase association from a configuration project OR delete a standalone (autonomous) server application. Destructive: guarded by a confirm-pre… |
delete_project | Remove an EDT project from the workspace, optionally deleting its files from disk (deleteContent). Destructive: guarded by a confirm-preview - call without c… |
export_configuration_to_xml | Export an EDT configuration project to XML files (EDT menu: Export -> Configuration to XML Files). Equivalent of 1C platform DumpConfigToFiles. |
get_check_description | Get detailed description of an EDT check by its ID. Returns markdown content with check explanation, examples, and how to fix. Accepts the symbolic check id… |
get_event_log | Read a 1C infobase event log WITHOUT a running 1C session by parsing the raw log files (legacy text ver 2.0: a 1Cv8.lgf dictionary + dated *.lgp partitions).… |
get_markers | List workspace markers: bookmarks and/or task markers (TODO, FIXME, XXX, HACK). Filter by markerKind (bookmark | task; omit to list both), projectName, fileP… |
get_mcp_history | Return the recorded MCP call history (this server's in-memory ring of request/response exchanges) so you can introspect your OWN traffic: which tools you cal… |
get_platform_documentation | Look up 1C:Enterprise platform documentation for built-in types (ValueTable, Array, Structure) and global built-in functions, including their methods, proper… |
get_problem_summary | Get problem summary with counts grouped by project and EDT severity level (ERRORS, BLOCKER, CRITICAL, MAJOR, MINOR, TRIVIAL). Use this for severity totals on… |
get_project_errors | List EDT configuration problems (validation markers) with optional project / severity / check-id / object filters. Each row carries the check code, message,… |
import_configuration_from_xml | Import a configuration from a directory of XML files into a NEW EDT project (EDT menu: Import); the reverse of export_configuration_to_xml. The projectName m… |
list_git_branches | List a project's git branches: local and remote-tracking, with the CURRENT branch marked (detached HEAD flagged), plus the 1C application/infobase each branc… |
resync_to_disk | Bulk re-synchronize the in-memory BM model to the on-disk src/ .mdo files and report BM-to-disk desync. Direction: MODEL -> DISK (writes the model out to src… |
revalidate_objects | Revalidate EDT project or specific objects. If objects array is empty or missing, revalidates entire project. FQN examples: 'Document.SalesOrder', 'Catalog.P… |
set_branch_infobase | Attach or detach an EXISTING infobase (application) to/from a specific git branch context, so switch_git_branch's automatic binding follows that branch. Targ… |
set_infobase_credentials | Store infobase connection credentials (user/password) so update_database and debug_launch can authenticate the update agent on an infobase that has a user li… |
switch_git_branch | Switch a project's git repository to another branch (headless EGit checkout). branch may be a short local name (e.g. 'feature/x') or a full ref ('refs/heads/… |
update_database | Apply configuration changes to an application's database (infobase), full or incremental. Target by launchConfigurationName (preferred) or projectName + appl… |
validate_xdto_package | Validate a single XDTO package by running EDT's OWN configuration validation (the same check engine behind get_project_errors) scoped to that package, and re… |
Run raw git commands (status/diff/commit/push/pull/...) in a project's repository via the 'git' tool. Powerful (it can push, checkout, stash); DISABLED by default - check it in the MCP Server Tools preference tab to enable.
| Tool | Description |
|---|---|
git | Run a git command in a project's repository - the non-UI equivalent of typing it in a terminal. Send it as a shell-style string (e.g. 'status', 'diff HEAD~1'… (not enabled by default) |
1C XDTO packages (XDTOPackage) can be authored end-to-end through MCP, down to their
package-local structure, without hand-editing XML:
create_metadata / modify_metadata / delete_metadata address XDTO
package members by FQN, layered on top of the existing top-level XDTOPackage.
create/delete:
XDTOPackage..ObjectType. — an ObjectType in the package (optional
flags: open, abstract, mixed, ordered, sequenced).XDTOPackage..Property. — a package-global Property.XDTOPackage..ObjectType..Property. — a Property nested in an
ObjectType.A Property's type accepts a built-in XSD type name (e.g. string), the exact name
of another ObjectType already in the same package (a same-package reference), or an
explicit {nsUri, name} pair; optional lowerBound/upperBound, nillable,
fixed+default round out the vocabulary. See get_tool_guide('create_metadata')
for the full parameter list.
validate_xdto_package runs EDT's own configuration validation scoped to one
package and returns a pass/fail verdict plus any problems found (e.g. a Property left
referencing an ObjectType that was since deleted) — a thin, read-only wrapper over
get_project_errors, handy right after authoring or editing package members.
Each tool declares a response type (IMcpTool.getResponseType()). The default — and the most common type — is Markdown: any tool that does not override getResponseType() returns Markdown as an EmbeddedResource with mimeType: text/markdown. The non-default types are enumerated exhaustively below; everything not listed here is Markdown.
structuredContent/JSON exists for clients to consume, not for the agent to read: it is justified only by verbatim round-trip of identifiers, machine-structured positions, a declared outputSchema, or UI-rendered data. Markdown is the default because it is more token-efficient and directly readable.
A tool returns JSON only when its result carries at least one of:
outputSchema;An action/confirmation/status result with none of these returns Markdown. write_module_source is the reference Markdown action tool; revalidate_objects, export_configuration_to_xml, and import_configuration_from_xml follow it (status + paths/counts, no round-trip data).
Which tool families stay JSON, and why:
create_metadata, modify_metadata, delete_metadata, via AbstractMetadataWriteTool) — return the edited object's round-trip FQN (a);validate_query — returns the error line/column (b);list_configurations — returns config identities consumed by other tools (a);clean_project / update_database — return a destructive status whose JSON shape is asserted by e2e (d-like contract).Errors are reported the same way regardless of a tool's normal format — see the Error contract below.
mimeType: text/markdown. This includes all read/list/search/navigation tools that emit human-readable reports — for example list_projects (which switches to JSON when called with format='json'), list_modules, list_subsystems, list_configurations, get_project_errors, validate_xdto_package, get_markers, get_problem_summary, get_check_description, get_metadata_objects, get_metadata_details, get_module_structure, get_subsystem_content, get_symbol_info, get_method_call_hierarchy, get_objects_by_tags, get_tags, get_platform_documentation, find_references, go_to_definition, search_in_code, read_module_source, read_method_source, write_module_source, rename_metadata_object, run_yaxunit_tests, terminate_launch, revalidate_objects, export_configuration_to_xml, import_configuration_from_xml, and all three LanguageTool tools (generate_translation_strings, translate_configuration, get_translation_project_info). (list_configurations is the exception among the list_* tools — it returns JSON; see below.)get_configuration_properties — returns a human-readable YAML body as an EmbeddedResource (resource named *.yaml, mimeType: text/yaml).structuredContent): get_server_status, get_applications, create_infobase, delete_infobase, get_content_assist, get_variables, get_profiling_results, list_configurations, list_breakpoints, set_breakpoint, remove_breakpoint, step, resume, wait_for_break, debug_launch, debug_status, debug_yaxunit_tests, evaluate_expression, start_profiling, stop_profiling, validate_query, clean_project, update_database, delete_project, git, plus the metadata-write tools that inherit JSON from AbstractMetadataWriteTool (create_metadata, modify_metadata, delete_metadata).get_edt_version, get_form_layout_snapshot.get_form_screenshot — returns the rendered form as an EmbeddedResource with an image/* mimeType.Whatever a tool's normal output format above, it reports a failure the same way: a JSON payload {"success": false, "error": ""} that the server delivers as a structured tool error (isError: true) regardless of the declared response type. The error field is always present (a null exception message is coalesced to Unknown error) and the message carries no redundant Error: prefix. A client detects failure via isError / success: false and reads error for the reason — no markdown parsing required. Success and purely informational results (for example "No references found", or a not-found accompanied by a list of valid alternatives) stay in the tool's natural format.
Every tool in the tools/list response carries an annotations object with the standard MCP behavioral hints, so a client can reason about a tool's effect before calling it. The hints are derived centrally from the tool name (a tool may override them):
| Hint | Meaning | When set |
|---|---|---|
readOnlyHint | The tool does not modify the workspace | true for get_* / list_* / read_* / search_* / find_* / validate_*; false for write and destructive tools |
idempotentHint | Repeating the call has no additional effect | true for the read-only tools above |
destructiveHint | The tool may perform a destructive or irreversible update | true for delete_metadata, clean_project, update_database, rename_metadata_object, import_configuration_from_xml, git |
openWorldHint | The tool interacts with an external/open world | false for every tool but git, which sets it true: push / pull / fetch reach a remote |
Only hints that apply are emitted; unset hints are omitted from the JSON. Tools that write but are not destructive (for example write_module_source, create_metadata) carry readOnlyHint: false and destructiveHint: false.
| Endpoint | Method | Description |
|---|---|---|
/mcp | POST | MCP JSON-RPC (initialize, tools/list, tools/call) |
/mcp | GET | Server info |
/health | GET | Health check |
The MCP server is a local developer tool and is secured for that model:
127.0.0.1 only. To expose it on all interfaces, enable Allow remote (non-loopback) access in MCP preferences — and set an auth token when you do.Authorization: Bearer (scheme case-insensitive, or the raw token) on every /mcp request. An empty token disables authentication (the default). /health is always unauthenticated (liveness only).evaluate_expression (runs arbitrary BSL in the running 1C app during a debug session) and destructive tools (update_database, clean_project, delete_metadata, rename_metadata_object). Treat any client that can reach the endpoint as fully trusted.export_configuration_to_xml / import_configuration_from_xml / build_external_objects read or write arbitrary filesystem paths (the broadest FS primitives in the surface; build_external_objects writes compiled .epf/.erf to a caller-chosen directory). They are trusted-caller-only; a warning is logged and the result flags outsideWorkspace when a path is outside the EDT workspace.Before a destructive metadata write, the server can ask you (the human at the EDT
workbench) to confirm — so the AI cannot silently delete, rename or retype configuration objects.
The gated tools are delete_metadata, rename_metadata_object, delete_project,
delete_infobase, update_database, and modify_metadata only when it changes an object's or
attribute's data type (a benign property edit is never gated).
Configure it in Window → Preferences → MCP Server:
Automation / CI bypass. Because the confirmation dialog would block a headless or automated run,
set the environment variable EDT_MCP_DESTRUCTIVE_CONSENT=allow on the EDT process before launch —
it overrides the preference and lets every gated operation proceed without a dialog (the same knob the
e2e suite uses). When there is no active workbench window (a headless server), the gate never blocks
either. The dialog only ever appears on a live UI session at the Ask level.
The prompt is time-bounded. A confirmation dialog waits at most 120 seconds for a human to
answer (below common MCP client request budgets, so a caller gets an actionable error instead of its
own transport timing out first). If nobody answers in time, the operation auto-rejects — nothing is
changed — with an error naming the tool, the 120 s budget, and the three ways to proceed: allow the
tool via Preferences (Allow all or per-tool), set EDT_MCP_DESTRUCTIVE_CONSENT=allow for unattended
runs, or re-run the call and answer the dialog promptly.
When a target infobase has a user list, connecting to it during update_database or debug_launch
can raise 1C's blocking "Configure Infobase access Settings" login dialog. To keep unattended MCP
calls from hanging on it, the server auto-cancels that dialog while a tool is running — the
MCP-triggered connect fails fast with a hint to set_infobase_credentials instead of blocking forever
(issue #194). Store the credentials once with set_infobase_credentials and the connect no longer
needs the dialog.
The auto-cancel is activity-scoped: it fires only while an MCP tool is in flight (plus a short grace window for the asynchronous read-back that follows a tool). So when the MCP server is idle, you can open EDT's "Configure Infobase access" dialog by hand in the GUI and use it normally — the credentials you enter are stored into EDT's encrypted Secure Storage (the leak-free path), exactly as the configurator does it. Only dialogs raised by MCP activity are cancelled; a human configuring credentials between agent runs is never interrupted. (The Secure-Storage password-hint dialog stays suppressed unconditionally — it is internal and never human-configured.)
Automation / CI bypass. The auto-cancel is on by default. To turn it off — e.g. to debug the
login flow interactively, or on a stand where you want the prompt to appear even during MCP calls — set
the environment variable EDT_MCP_SUPPRESS_AUTH_DIALOG=false (also 0 / no) on the EDT process
before launch. Any other value — or leaving it unset — keeps auto-cancel enabled, so an unattended run
never regresses into a hang.
Organize your metadata objects with custom tags for easier navigation and filtering.
Tags help you:
Assigning Tags to Objects:
Managing Tags:
In the Manage Tags dialog you can:
Tagged objects show their tags as a suffix in the Navigator tree:
To enable/disable tag display:
Filter the entire Navigator to show only objects with specific tags:
The Navigator will show only:
To clear the filter: Click Turn Off in the dialog or use the toolbar button again.
Quickly toggle tags on selected objects using keyboard shortcuts:
| Shortcut | Action |
|---|---|
| Ctrl+Alt+1 | Toggle 1st tag |
| Ctrl+Alt+2 | Toggle 2nd tag |
| ... | ... |
| Ctrl+Alt+9 | Toggle 9th tag |
| Ctrl+Alt+0 | Toggle 10th tag |
Features:
To customize shortcuts: Window → Preferences → General → Keys → search for "Toggle Tag"
Find metadata objects that haven't been tagged yet:
The Navigator will show only objects that have no tags assigned, making it easy to identify objects that need categorization.
Assign or remove tags from multiple objects at once:
Behavior:
For advanced filtering across multiple projects, use the Tag Filter View:
Window → Show View → Other → MCP Server → Tag Filter
This view provides:
Tags are stored in .settings/metadata-tags.yaml file in each project. This file:
Example:
assignments:
CommonModule.Utils:
- Utils
Document.SalesOrder:
- Important
- Sales
tags:
- color: '#FF0000'
description: Critical business logic
name: Important
- color: '#00FF00'
description: ''
name: Utils
- color: '#0066FF'
description: Sales department documents
name: Sales
Organize your Navigator tree with custom groups to create a logical folder structure for metadata objects.
Groups help you:
Creating a Group:
Create Group Dialog:
Adding Objects to a Group:
Removing Objects from a Group:
Grouped objects appear inside their group folders in the Navigator tree:
Key Features:
| Action | How to Do It |
|---|---|
| Create group | Right-click folder → New Group... |
| Add object to group | Right-click object → Add to Group... |
| Remove from group | Right-click object in group → Remove from Group |
| Copy group name | Select group → Ctrl+C |
| Delete group | Right-click group → Delete |
| Rename group | Right-click group → Rename... |
| Hide/show groups | Click Hide Groups in the Navigator toolbar |
Groups are stored in .settings/groups.yaml file in each project. This file:
Example:
groups:
- name: "Products & Inventory"
description: "Product and inventory catalogs"
path: Catalog
order: 0
children:
- Catalog.ItemKeys
- Catalog.Items
- Catalog.ItemSegments
- Catalog.Units
- Catalog.UnitsOfMeasurement
- name: "Organization"
description: "Organization structure catalogs"
path: Catalog
order: 1
children:
- Catalog.Companies
- Catalog.Stores
- name: "Core Functions"
description: "Core shared functions used across the application"
path: CommonModule
order: 0
children:
- CommonModule.CommonFunctionsClient
- CommonModule.CommonFunctionsServer
- CommonModule.CommonFunctionsClientServer
- name: "Localization"
description: "Multi-language support modules"
path: CommonModule
order: 1
children:
- CommonModule.Localization
- CommonModule.LocalizationClient
- CommonModule.LocalizationServer
- CommonModule.LocalizationReuse
The plugin is a Maven/Tycho project under mcp/. CI builds it via .github/workflows/build.yml; the same flow can be run locally with source/compile.sh.
mvnw wrapper is committed — install Maven manually or via a package manager: winget, Homebrew, apt, SDKMAN, etc.)bash (Git Bash on Windows works) and either zip or the jar binary that ships with the JDKhttps://edt.1c.ru/, https://download.eclipse.org/ and Maven Central — Tycho downloads the EDT p2 repository and Eclipse SDK on the first run (hundreds of MB, cached afterwards under ~/.m2/)# from the repo root
bash source/compile.sh --skip-tests
Output:
source/dist/MCP-EDT.v.zip
This is a valid p2 update site — install via EDT → Help → Install New Software → Add → Archive….
source/compile.sh accepts every path as a flag (with matching environment-variable fallback) so it can be driven from CI or run against an out-of-tree checkout:
| Flag | ENV fallback | Default | Meaning |
|---|---|---|---|
--skip-tests | — | off | Skip Maven Surefire tests |
--version X.Y.Z | — | parsed from README.md, falls back to dev | Version label used in the output zip name |
--archive-prefix PREFIX | — | MCP-EDT.v | Archive name prefix (final name: .zip) |
--project-root PATH | EDT_MCP_PROJECT_ROOT | parent of script dir | Repo root containing mcp/ |
--mcp-dir PATH | — | /mcp | Maven project directory |
--repo-dir PATH | — | /mcp/repositories/com.ditrix.edt.mcp.server.repository/target/repository | Tycho p2 output to repackage |
--output-dir PATH | EDT_MCP_OUTPUT_DIR | /dist | Where the final zip lands |
--java-home PATH | JAVA_HOME | — | JDK 17 home; if set, prepended to PATH for Maven |
--maven-home PATH | MAVEN_HOME / M2_HOME | — | Maven home (uses /bin/mvn); otherwise falls back to mvn on PATH |
-h, --help | — | — | Show help |
# Self-contained invocation, no env tweaks required
bash source/compile.sh \
--java-home "/c/Program Files/Java/jdk-17" \
--maven-home /d/Soft/maven \
--skip-tests \
--version 1.27.1
# Drop the artifact somewhere else
bash source/compile.sh --output-dir /tmp/edt-mcp-builds
# Same, configured via environment
JAVA_HOME="/c/Program Files/Java/jdk-17" \
MAVEN_HOME=/d/Soft/maven \
EDT_MCP_OUTPUT_DIR=/tmp/edt-mcp-builds \
bash source/compile.sh
mcp/targets/default/default.target) and the Eclipse 2023-12 release — expect several minutes. Subsequent builds run in ~1 minute thanks to the local p2 cache.jar when zip is unavailable) so it installs cleanly on both Windows and Linux EDT instances.source/dist/ is gitignored; only the script itself is tracked.edt-mcp-autopilot skillThis repository ships a Claude Code project skill at .claude/skills/edt-mcp-autopilot/ that takes a whole task or issue end-to-end with minimal human input, using a multi-agent, Spec-Driven (SDD) pipeline. The code-conduct it follows lives in CLAUDE.md and the sibling skills under .claude/skills/.
Pipeline (phases): study the task → documentation researchers → several waves of code researchers (loop-until-dry) → adversarial critics (refuted findings dropped, "rework" bounced back) → an architect that synthesises a spec and a file-disjoint developer partition → N parallel developers → a 3–4 reviewer loop until clean → build + unit tests → live-stand scenarios → a Russian issue comment and a pull request.
How to run (Claude Code): the skill is auto-discovered from .claude/skills/. Invoke it with
/edt-mcp-autopilot
It runs unattended: the only mandatory human touchpoint is confirming the live-stand result before shipping; principal design questions are posted to the issue and polled until answered. The heavy fan-out runs through the Claude Code Workflow tool (the two scripts under references/), orchestrated by the main session as a "conductor". Scale the fan-out with the size argument (small / medium / large).
It deliberately spawns many subagents — documentation and code researchers in waves, a critic panel, parallel developers, and a cyclic reviewer pool. Measured cost per real task in this repository (subagent tokens only — the conductor, the Maven build and the live-stand verification add more on top):
| Phase | Subagents | Subagent tokens | Wall time |
|---|---|---|---|
| Discover (research → critics → architect) | ~8–9 | ~0.6–0.9 M | ~15 min |
| Build (parallel developers → reviewer loop) | ~11–15 | ~0.7–0.8 M | ~15 min |
So a single task's discover + build phases alone are roughly 20–25 subagents and ~1.3–1.8 M subagent tokens; a full end-to-end task (with the conductor, the builds, golden regeneration and live-stand checks) typically runs several million tokens and 30–60+ minutes. A misconfigured run can be far worse — an early, pre-hardened version once burned ~9.8 M tokens / ~210 agents on a no-op before the guard rails were added.
Use it for substantial whole-task work — a feature, a non-trivial bug, or a new tool — not for small edits or quick questions, where working directly is far cheaper.
Install via CLI
npx mdskills install DitriXNew/edt-mcpEDT MCP Server is a free, open-source AI agent skill. MCP (Model Context Protocol) server plugin for 1C:EDT, enabling AI assistants (Claude, GitHub Copilot, Cursor, etc.) to interact with EDT workspace. - MCP Protocol 2025-11-25 - Streamable HTTP transport with SSE support - Project Information - List workspace projects and configuration properties - Error Reporting - Get errors, warnings, problem summaries with filters - Check Descriptio
Install EDT MCP Server with a single command:
npx mdskills install DitriXNew/edt-mcpThis downloads the skill files into your project and your AI agent picks them up automatically.
EDT MCP Server works with Claude Code, Claude Desktop, Cursor, Vscode Copilot, Windsurf, Continue Dev, Gemini Cli, Amp, Roo Code, Goose. Skills use the open SKILL.md format which is compatible with any AI coding agent that reads markdown instructions.