{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_ppymvtv7dd4m","handle":"ml-training-init-mcp","url":"https://wellknown.network/agents/ml-training-init-mcp","links":{"self":"https://wellknown.network/agents/ml-training-init-mcp/record.json","html":"https://wellknown.network/agents/ml-training-init-mcp","markdown":"https://wellknown.network/agents/ml-training-init-mcp/record.md","api":"https://wellknown.network/api/v1/agents/ml-training-init-mcp","status":"https://wellknown.network/api/v1/agents/ml-training-init-mcp/status","claim":"https://wellknown.network/agents/ml-training-init-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/ml-training-init-mcp/claim.json","badge":"https://wellknown.network/agents/ml-training-init-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/ml-training-init-mcp/history","tools":"https://wellknown.network/api/v1/agents/ml-training-init-mcp/tools"},"ard":{"identifier":"urn:air::server:ml-training-init-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"ml-training-init-mcp","summary":"MCP server for ML training script initialization with file locking","description":"# ML Training Init MCP Server\n\nA Model Context Protocol (MCP) server that implements a sequential thinking pattern for ML training script generation with managed file constraints.\n\n## Key Feature: Managed File System (1 Training + 2 Configs)\n\n**IMPORTANT**: This server enforces a managed file constraint system:\n- **1 Training Script**: The main ML training file that will be executed\n- **Up to 2 Config Files**: Optional configuration files (YAML, JSON, .env, etc.)\n- ALL operations work only with these managed files\n- Prevents agents from creating unnecessary files when stuck\n- Clearly identifies ML training workflows with [ML TRAINING WORKFLOW] tags\n\n## Installation\n\n```bash\npip install -r requirements.txt\n```\n\n## Running the Server\n\n```bash\npython -m src.server\n```\n\n## Usage with Claude Desktop\n\nAdd to your Claude Desktop configuration:\n\n```json\n{\n  \"mcpServers\": {\n    \"ml-training-init\": {\n      \"command\": \"python\",\n      \"args\": [\"-m\", \"src.server\"],\n      \"cwd\": \"/path/to/ml-training-init-mcp\"\n    }\n  }\n}\n```\n\n## Available Tools\n\n### 1. `initialize_training_file`\n- Creates the main training script file\n- Takes: file_name, content, reference\n- Returns: file_path, file_type, managed_files status\n- AI agents use this for ML training workflows\n\n### 2. `create_config_file`\n- Creates configuration files (max 2 allowed)\n- Supports: YAML, JSON, TOML, .env, .ini, etc.\n- Takes: file_name, content, config_type\n- Returns: file_path, file_type, managed_files status\n\n### 3. `get_managed_files`\n- Lists all managed files (training + configs)\n- Shows file paths and names\n- Returns: training file info, config files list\n\n### 4. `get_file_content`\n- Get content of a specific managed file\n- Takes: file_name\n- Returns: file_path, content, file_type\n\n### 5. `get_current_file`\n- Quick access to training file content\n- Returns: file_path, content\n\n### 6. `monitor_and_fix`\n- Fix errors in any managed file\n- Takes: error_trace, file_name (optional)\n- Auto-parses common Python/ML errors\n\n…","publisher":{"name":"ML Training Init","url":null},"homepage":null,"repository":null,"version":"0.2.2","license":"MIT","protocols":["mcp"],"tags":["mcp"],"pricing":null,"endpoints":[{"url":"pypi:ml-training-init-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","license":"pypi","summary":"pypi","version":"pypi","description":"pypi","publisherName":"pypi"}},"derived":{"capabilities":[{"slug":"dev.filesystem","name":"Filesystem","confidence":0.825,"provenance":"derived"}],"categories":["dev"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"ml-training-init-mcp","registry":"pypi","observedAt":"2026-11-03T18:22:58.324Z","publishedAt":"2025-08-27T04:28:05.674570Z","latestVersion":"0.2.2"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"ml-training-init-mcp","url":"https://pypi.org/project/ml-training-init-mcp/","firstSeenAt":"2026-09-10T06:23:24.878Z","fetchedAt":"2026-11-03T18:21:35.995Z","normalizedAt":"2026-11-03T18:21:35.995Z"}]},"firstSeenAt":"2026-09-10T06:23:24.878Z","updatedAt":"2026-11-03T18:22:58.324Z"}