Skip to main content
Glama

OpenRouter Agents MCP Server

by wheattoast11
CHANGELOG.md4.58 kB
# Changelog ## [1.5.0](https://github.com/wheattoast11/openrouter-deep-research-mcp/compare/v1.4.0...v1.5.0) (2025-08-26) ### Features * 2025 upgrades — model catalog 2025 prefs, kurtosis-style ensembles + multimodal fallbacks; PGlite adaptive thresholds + keyword fallback + HNSW params; MCP HTTP auth OAuth2/JWT scaffolding + CORS; planning AIMD; OpenRouter hybrid batching; gen-docs script and CHANGELOG/README hooks ([a8a18aa](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/a8a18aa489236c01d14b0e5c5b5e6ad71e52a56c)) * compact prompts/resources, hybrid index tools, planning fallbacks, strict citation prompts, compact tool params, README v1.2 with architecture & journeys ([0e81d33](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/0e81d3387c0d82eca647a23b2869dbe31db6ee18)) * **docs:** update README to include new AGENT and MANUAL modes, enhance tool descriptions, and add npm/GitHub Packages publishing details ([8fdad28](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/8fdad28712ce403dad7f71372f660e4e3519894e)) * enhance configuration and functionality of MCP server tools; introduce unified retrieve schema for hybrid search and SQL execution; improve job status reporting with concise summaries; update research history listing for better readability and filtering; refine DB initialization logic for improved retry handling ([f34f997](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/f34f9979a87c76b25f640e3f0a9f910e8a64998b)) * enhance model configuration with new high and low-cost models; implement advanced caching and cost optimization strategies; update README with model details and caching configurations ([100f818](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/100f818b07f567e84aa84c5cfdd156b49d74c196)) * implement async job processing with submit, status, and cancel functionalities; enhance configuration for HTTPS and reranking; update README with new demo scripts and MCP client JSON configuration examples ([70e66a2](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/70e66a274a5a99cb6cf7834a8b1c58cad6538e75)) * lightweight web tools (search_web, fetch_url) and docs; register tools ([1cdb005](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/1cdb0059f653bb7d6e62da9d4d124bd06c18de11)) * **mode:** AGENT/MANUAL/ALL w/ always-on tools; add agent+ping; async submit UI/SSE links; docs/CI; v1.3.2 ([17b456e](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/17b456ef014b2d649a4bac88e6b3e7999c5d1188)) * **models:** switch defaults to anthropic/claude-sonnet-4 and openai/gpt-5 family ([3348bd8](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/3348bd846a38b7864ce619518f0f8cb30eb206dc)) * streaming robustness, dynamic catalog tools, DB QoL tools, README updates (Aug 09, 2025) ([65cf942](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/65cf942c8db3756f6df5a52308b5a02b46f73b3d)) * tarball backups, refreshed default model IDs, extended tests, docs update; npm audit fix ([8918fcc](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/8918fcc1194511f8d55c3ed19aa6ee4b1c80ecdc)) * update model configurations to include new planning defaults and enhance research agent capabilities; add client context handling and hyper mode for optimized model selection; improve web scraping with additional search strategies and usage tracking ([2850f59](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/2850f59033106a4ecd8fb5c683f92920ff33ff96)) ### Bug Fixes * define search_web implementation ([5eef4a4](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/5eef4a41da07d239bab54b4f809cc6a50991a92f)) * increase classifier max_tokens to satisfy provider minimums; docs: changelog v1.2.0 ([d469cea](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/d469ceac67462fbf39a75f472b5c6b8ca8ccdead)) ### Performance Improvements * bounded parallelism for research; dynamic vision capability via catalog; add PARALLELISM config ([248046d](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/248046d8b11aa2112eb5f9ef2848b94581562fe2)) ## [1.4.0](https://github.com/wheattoast11/openrouter-deep-research-mcp/compare/v1.3.1...v1.4.0) (2025-08-26) ### Features * **mode:** AGENT/MANUAL/ALL w/ always-on tools; add agent+ping; async submit UI/SSE links; docs/CI; v1.3.2 ([17b456e](https://github.com/wheattoast11/openrouter-deep-research-mcp/commit/17b456ef014b2d649a4bac88e6b3e7999c5d1188))

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/wheattoast11/openrouter-deep-research-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server