Skip to main content

MCP server for logging Vise Coding sessions.

Project description

Vise Logger

Vise Logger is an MCP server that captures, rates, and archives Vise Coding sessions from various AI coding tools (Cursor, GitHub Copilot, Cline, ...), storing them at viselo.gr for reference, comparisons, and searchability.

As Sean Grove (OpenAI) and Dexter Horthy (HumanLayer) say: The source spec (i.e. coding sessions) is the valuable artifact.

Just like you used to store the source code, not binaries, in GitHub, you should now store your coding sessions as main source of truth.

On viselo.gr, you have

  • an overview of your sessions
  • a backup, as you can download each session as zip, containing the original format
  • can share sessions with others
  • get insights from your sessions and of shared sessions.

Features planned for the future: search and statistics (see https://fb-swt.gi.de/fileadmin/FB/SWT/Softwaretechnik-Trends/Verzeichnis/Band_45_Heft_3/5_TAV51_Farago.pdf).

Installation and Setup

Local installation

This project uses uv for package and environment management.

  1. Create a virtual environment:

    uv venv
    
  2. Activate the virtual environment:

    • On macOS/Linux:
      source .venv/bin/activate
      
    • On Windows:
      .venv\Scripts\activate
      
  3. Install the project in editable mode: This command installs the project and its dependencies into the virtual environment. The -e flag (editable) ensures that any changes you make to the source code are immediately available without needing to reinstall.

    uv sync
    

    This installs the runtime dependencies, in editable mode. To additionally install the dev dependencies, add argument --extra dev

Installation of MCP Server via PyPI

Making use of the MCP server in your AI coding tool

In Cline and Roo:

"vise-logger": {
  "disabled": false,
  "timeout": 60,
  "type": "stdio",
  "command": "uv",
  "args": [
    "--directory",
    "/home/emergency/git/vise-logger",
    "run",
    "vise-logger"
  ],
  "env": {
    "VISE_LOG_API_KEY": "your API key from viselo.gr",
    "OTEL_RESOURCE_ATTRIBUTES": "service.name=vise-logger,service.namespace=public-logs,deployment.environment=production",
    "OTEL_EXPORTER_OTLP_ENDPOINT": "https://otlp-gateway-prod-eu-west-2.grafana.net/otlp",
    "OTEL_EXPORTER_OTLP_HEADERS": "Basic MTM2MjQxMTpnbGNfZXlKdklqb2lNVFV5TWpJM05pSXNJbTRpT2lKMmFYTmxMV3h2WjJkbGNpSXNJbXNpT2lKMGNtd3ljRGM1U0ZCU1ZWQjNORGRrZEV3M05EUTNORTRpTENKdElqcDdJbklpT2lKd2NtOWtMV1YxTFhkbGMzUXRNaUo5ZlE9PQ=="
  }
}

Environment Variables

  • VISE_LOG_API_KEY: Required. The API key for authenticating with the Vise Logger backend. This key is necessary for uploading sessions. Get it at https://viselo.gr/ by signing in, going to settings and then "Generate New Key". The key is only shown once, so make sure you save it.

  • OTEL_RESOURCE_ATTRIBUTES (default "service.name=vise-logger,service.namespace=public-logs,deployment.environment=production"), OTEL_EXPORTER_OTLP_ENDPOINT (default "https://otlp-gateway-prod-eu-west-2.grafana.net/otlp"), and OTEL_EXPORTER_OTLP_HEADERS (default "Basic MTM2MjQxMTpnbGNfZXlKdklqb2lNVFV5TWpJM05pSXNJbTRpT2lKMmFYTmxMV3h2WjJkbGNpSXNJbXNpT2lKMGNtd3ljRGM1U0ZCU1ZWQjNORGRrZEV3M05EUTNORTRpTENKdElqcDdJbklpT2lKd2NtOWtMV1YxTFhkbGMzUXRNaUo5ZlE9PQ=="): Optional. The logs are sent to the MCP host and to a file (see below), but you can also use open telemetry to send your logs to a server with an open telemetry endpoint. If you use the default values, they are sent to the developer of this project.

Running the MCP Server

To run the MCP server directly from the command line:

vise-logger

MCP Tools

This server provides two tools:

rate_and_upload(stars: float, comment: str = "")

Rates and archives the current Vise Coding session. It finds the relevant log file, filters it for privacy, and simulates an upload to the backend.

configure_log_dir()

Scans the system to find and verify the directory where your coding tool saves its logs. This helps speed up future searches.

Warning: This can be a very long-running operation, potentially taking minutes to hours, as it may scan your entire hard drive.

Running Tests

With the virtual environment activated and the project installed in editable mode, you can run the integration tests:

python3 -m unittest discover tests

Note on Best Practices: Installing the package in editable mode (-e) is the recommended way to run tests for a distributable Python package. It correctly resolves imports without needing to modify sys.path, which is a less robust method. This approach simulates a real installation, making the testing environment more realistic.

REST Endpoints The Sessions Are Sent to

The endpoint should have Content-Type: multipart/form-data and the following form fields:

  • file (required): The zip file containing the AI coding session data
  • metadata (required): JSON string with AI coding session metadata. The metadata JSON structure:
json{
  "marker": "string (required)",
  "tool": "string (required)",
  "stars": "number (required)",
  "comment": "string (optional)"
}

The official web application for Vise Logger is www.viselo.gr. You can test it via

curl -X POST \
  -F "file=@session.zip" \
  -F 'metadata={"marker": "Rated session at 2025-07-30-20-56-11: 1.9 stars.", "tool": "curl", "stars": 1.9, "comment": "Just a curl test"}' \
  "https://studio--viselog.us-central1.hosted.app/api/v1/sessions"

MCP server's logging and debugging

MCP-Server's own logs: if environment variables OTEL_RESOURCE_ATTRIBUTES, OTEL_EXPORTER_OTLP_ENDPOINT, and OTEL_EXPORTER_OTLP_HEADERS are set, open telemetry (http/protobuf) is used for logging. You can log to the developer's cloud instance using

OTEL_RESOURCE_ATTRIBUTES="service.name=vise-logger,service.namespace=public-logs,deployment.environment=production"
OTEL_EXPORTER_OTLP_ENDPOINT="https://otlp-gateway-prod-eu-west-2.grafana.net/otlp"
OTEL_EXPORTER_OTLP_HEADERS="Basic MTM2MjQxMTpnbGNfZXlKdklqb2lNVFV5TWpJM05pSXNJbTRpT2lKMmFYTmxMV3h2WjJkbGNpSXNJbXNpT2lKMGNtd3ljRGM1U0ZCU1ZWQjNORGRrZEV3M05EUTNORTRpTENKdElqcDdJbklpT2lKd2NtOWtMV1YxTFhkbGMzUXRNaUo5ZlE9PQ=="

Logs are also written to a file, e.g. ~/.local/state/vise-logger/log/mcp_server.log (path depends on your machine, see https://pypi.org/project/platformdirs/).

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

vise_logger-0.1.0.tar.gz (1.5 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

vise_logger-0.1.0-py3-none-any.whl (636.5 kB view details)

Uploaded Python 3

File details

Details for the file vise_logger-0.1.0.tar.gz.

File metadata

  • Download URL: vise_logger-0.1.0.tar.gz
  • Upload date:
  • Size: 1.5 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.7.8

File hashes

Hashes for vise_logger-0.1.0.tar.gz
Algorithm Hash digest
SHA256 f930efffc9fdf03f0e057dd3ae8349bfb481d9c2f7b05b8127896117246a89bc
MD5 7c91e8dc55c2f9931475352a38ec5820
BLAKE2b-256 f5dee12b74d24caaf6a8b37af2d142f71ab398c44086f5b334f2f10607b14766

See more details on using hashes here.

File details

Details for the file vise_logger-0.1.0-py3-none-any.whl.

File metadata

File hashes

Hashes for vise_logger-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 db02d5ec2efcdee069b48ba73b4d73a6b6840f2524a59fc80fae5ca83c5e7df3
MD5 080a24da0fda67105e32262cade8d736
BLAKE2b-256 0592bc2cd55e1bfea0e90ce2b99193d2e1f4a16e9d2951aa6ec8299bd88ead2c

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page