scopix
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
Scopix
Scopix picks the code an AI model needs for a task. Give it a task and a token budget, and it returns the most useful files, functions and line ranges, with a reason for each one. It runs on your machine with no account or cloud service. Local AI is optional and experimental.
Install
Scopix needs Python 3.11 or later.
pip install scopixRelated MCP server: Sippion
Quick start
cd your-repo
scopix init
scopix select "fix the authentication refresh logic" --budget 6000
scopix export "fix the authentication refresh logic" --budget 6000 --output context.mdinit creates .scopix/ and indexes the repository. select prints the context,
and export saves it. Each result looks like this:
### 1. `src/payments/retry-policy.ts` (lines 1-61, ~469 tokens)
**Why:** imported by tests/payment-retry.test.ts (distance 1); defines
`RetryPolicy.shouldRetryResult` which matches the task; matches task terms
`retry`, `payment`, `attempt`Use with coding agents
Scopix can run as an MCP server, so agents such as Claude Code can ask it for context. See the setup guide for other clients.
pip install "scopix[mcp]"
claude mcp add scopix -- scopix serve --path /path/to/repoPython API
from scopix import Scopix
with Scopix(".") as engine:
engine.initialize()
engine.index() # only changed files are parsed after the first run
result = engine.select("add retry handling", token_budget=4000)
for item in result.items:
print(item.path, item.start_line, item.end_line, item.tokens)
print(result.to_markdown()) # or to_json(), to_text(), to_xml()How well it works
On 45 unseen commits from 9 projects, Scopix found 39% of the changed files in 2,000 tokens, while grep needed 16,000 tokens to reach 38%. It still misses about 4 in 10 at 8,000 tokens. See the benchmark.
Commands
Command | What it does |
| Create |
| Update the index |
| Print context for a task |
| Show why each item was chosen |
| Write context as Markdown, JSON, text or XML |
| Show index statistics |
| Show the dependency graph |
| Re-index when files change |
| Run an MCP server for coding agents |
| Compare results with simple baselines |
| Mark a file as helpful or not |
| Manage selection profiles |
| Set up an optional local AI model |
| Show or change settings |
Run scopix COMMAND --help for options, or see the reference.
Languages
Scopix parses TypeScript, JavaScript, Python, Go, Java, Rust, PHP and C#. It indexes other text files with less detail.
Privacy
By default, Scopix sends nothing over the network and never runs your code. It skips
files that match secret patterns such as .env and *.pem. Some optional features
use the network.
License
MIT. See LICENSE.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Deterministic context layer for your codebase: change impact, blast radius, answers with receipts.
The context platform for AI agents: code search, team decisions, and grounded project context.
Codebase intelligence for agents: 152 structured artifacts across 21 programs, one call.
Codebase intelligence for AI agents — dead code, blast radius, ownership.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceEnables AI coding agents to automatically compile the most relevant slice of a repository into context for a given task.7 npm1MIT
- FlicenseNot gradedqualityBmaintenanceEnables AI coding agents to retrieve focused, bounded repository context locally and read-only, reducing token usage while never modifying the repository or executing its code.-
- AlicenseBqualityBmaintenanceEnables coding agents to retrieve local, token-budgeted, source-traceable context from a workspace, with file:line provenance and selection receipts for chosen spans.538 PyPIApache 2.0
- AlicenseNot gradedqualityAmaintenanceEnables AI coding agents and IDEs to query a local structural code graph and retrieve focused, token-bounded context on code entities, Git relationships, callers/callees, impact, paths, and recorded mistakes for more efficient sessions.24 npmMIT