BifrostSAFE
VSCode Extension with an MCP server that exposes semantic tools like Find Usages and Rename to LLMs
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
<img src="https://img.shields.io/visual-studio-marketplace/d/ConnorHallman.bifrost-mcp?label=VSCode%20Extension%20Downloads&cacheSeconds=3600" alt="VSCode Extension Downloads" width="250">
This VS Code extension provides a Model Context Protocol (MCP) server that exposes VSCode's powerful development tools and language features to AI tools. It enables advanced code navigation, analysis, and manipulation capabilities when using AI coding assistants that support the MCP protocol.
Table of Contents
- Features
- Installation/Usage
- Multi-Project Support
- Available Tools
- Available Commands
- Troubleshooting
- Contributing
- Debugging
- License
Features
- Language Server Integration: Access VSCode's language server capabilities for any supported language
- Code Navigation: Find references, definitions, implementations, and more
- Symbol Search: Search for symbols across your workspace
- Code Analysis: Get semantic tokens, document symbols, and type information
- Smart Selection: Use semantic selection ranges for intelligent code selection
- Code Actions: Access refactoring suggestions and quick fixes
- HTTP/SSE Server: Exposes language features over an MCP-compatible HTTP server
- AI Assistant Integration: Ready to work with AI assistants that support the MCP protocol
Usage
Installation
- Install the extension from the VS Code marketplace
- Install any language-specific extensions you need for your development
- Open your pro
8b51149ac987OBSERVED · 2026-10-06Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add bifrost-mcp -- npx -y [email protected]
{
"mcpServers": {
"bifrost-mcp": {
"command": "npx",
"args": [
"-y",
"[email protected]"
]
}
}
}Exposed tools (19)
17 read · 2 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
find_implementations | read | Discovers all concrete implementations of an interface, abstract class, or abstract method in the codebase. |
find_usages | read | Finds all references to a symbol at a specified location in code. This tool helps you identify where functions, variables, types, or other symbols are used throughout the codebase. |
get_call_hierarchy | read | Analyzes and visualizes the call relationships between functions and methods in the codebase. |
get_code_actions | read | Provides context-aware code actions and refactoring suggestions at a specified location. |
get_code_lens | read | Gets CodeLens information for a document, showing actionable contextual information inline with code |
get_completions | read | Provides intelligent code completion suggestions based on the current context and cursor position. |
get_declaration | read | Finds declarations of a symbol at a specified location. This helps in navigating to where symbols are declared, particularly useful for imported symbols. |
get_document_highlights | read | Finds all highlights of a symbol in a document. This is useful for highlighting all occurrences of a symbol within the current document. |
get_document_symbols | read | Analyzes and returns a hierarchical list of all symbols defined within a document. |
get_hover_info | read | Retrieves comprehensive information about a symbol when hovering over it in code. |
get_rename_locations | write | Identifies all locations that need to be updated when renaming a symbol. |
get_selection_range | read | Gets selection ranges for a position in a document. This helps in smart selection expansion based on semantic document structure. |
get_semantic_tokens | read | Provides detailed semantic token information for enhanced code understanding and highlighting. |
get_signature_help | read | Provides detailed information about function signatures as you type function calls. |
get_type_definition | read | Finds type definitions of a symbol at a specified location. This is particularly useful for finding the underlying type definitions of variables, interfaces, and classes. |
get_type_hierarchy | read | Analyzes and visualizes the inheritance and implementation relationships between types. |
get_workspace_symbols | read | Searches for symbols across the entire workspace. This is useful for finding symbols by name across all files. Especially useful for finding the file and positions of a symbol to use in other tools. |
go_to_definition | read | Navigates to the original definition of a symbol at a specified location in code. |
rename | write | Identifies all locations that need to be updated when renaming a symbol, and performs the renaming. |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (6)
sse
icon.txt
placeholder.txt
.vscode-test.mjs
.vscodeignore
@modelcontextprotocol/sdk, @types/cors, @types/express, cors, express, @types/mocha, @types/vscode, @typescript-eslint/eslint-plugin
Gates applied: no_behavioural_pass.
8b51149ac987full audit observations/trust-audit/mcp-server/biegehydra__bifrost.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-06 | 8b51149ac987 | SAFE | B | 89 | first audit |
Questions
What is the Bifrost MCP server?
VSCode Extension with an MCP server that exposes semantic tools like Find Usages and Rename to LLMs
What tools does Bifrost expose?
19 in total: 17 read-only, 2 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Bifrost safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does Bifrost need?
No credential environment variables were found in its source, so it appears to need none.
How does Bifrost run?
It speaks sse, so it runs as a service you connect to over the network. It is published on npm as bifrost-mcp at 0.0.14.
How current is this page?
The grade is for one exact copy of the source (8b51149ac987), read on 2026-10-06. The repository is watched and re-audited when it changes.