Skip to main content
Glama
mnguyen72

Agent Search MCP

by mnguyen72

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
MODENoTransport mode, e.g. http
PROXY_URLNoShared proxy URL used when USE_PROXY=true
USE_PROXYNoSet to true to use PROXY_URL for all engines
EXA_API_KEYNoOptional API key for the Exa Search provider
YDC_API_KEYNoOptional API key for the You.com Search provider
OUTPUT_STYLENoSet to compact to reduce response tokens
RERANK_TOP_KNoTop K results for semantic reranking
BOCHA_API_KEYNoOptional API key for the Bocha Web Search provider
BRAVE_API_KEYNoOptional API key for the Brave Search provider
ENABLED_TOOLSNoTool registration allowlist
DENIED_ENGINESNoEngine execution denylist; deny wins
DISABLED_TOOLSNoTool registration denylist; deny wins
SEMANTIC_DEDUPNoEnable optional semantic deduplication (uses Python and Model2Vec)
SERPER_API_KEYNoOptional API key for the Serper Google Search provider
SNIPPET_LENGTHNoMaximum length of result snippets
TAVILY_API_KEYNoOptional API key for the Tavily Search provider
WIBY_PROXY_URLNoExplicit proxy URL for Wiby
ALLOWED_ENGINESNoEngine execution allowlist
ALLOWED_ORIGINSNoAllowed Origin headers for browser requests in HTTP mode
DEDUP_THRESHOLDNoThreshold for semantic deduplication
HTTP_AUTH_TOKENNoBearer token required for HTTP mode authentication
SEMANTIC_RERANKNoEnable optional semantic reranking (uses Python and Model2Vec)
SOGOU_PROXY_URLNoExplicit proxy URL for Sogou
WIBY_PROXY_URLSNoUser-owned proxy pool for Wiby as a JSON array of 2-16 HTTP(S) proxy URLs
MAX_FULL_RESULTSNoMaximum number of results returned with full detail
MOJEEK_PROXY_URLNoExplicit proxy URL for Mojeek
SOGOU_PROXY_URLSNoUser-owned proxy pool for Sogou as a JSON array of 2-16 HTTP(S) proxy URLs
MOJEEK_PROXY_URLSNoUser-owned proxy pool for Mojeek as a JSON array of 2-16 HTTP(S) proxy URLs
PAID_ENGINE_ORDERNoSelects the first configured optional provider; not a quality claimbrave,exa,tavily,youcom,tencent_wsa,bocha,serper
SEARCH_CACHE_TTL_MSNoTime-to-live for persisted exact-result cache entries in milliseconds
TENCENT_WSA_API_KEYNoOptional API key for the Tencent Web Search API provider
DUCKDUCKGO_PROXY_URLNoExplicit proxy URL for DuckDuckGo
SEARCH_PROVIDER_MODENoDefault routing policy: free_first, quality_escalation, paid_first, or free_onlyfree_first
DUCKDUCKGO_PROXY_URLSNoUser-owned proxy pool for DuckDuckGo as a JSON array of 2-16 HTTP(S) proxy URLs
EVIDENCE_BUDGET_CHARSNoEvidence-character budget cap across the complete response1200
SEARCH_CACHE_DIRECTORYNoDirectory to persist the exact-result cache locally
SEARCH_BUDGET_MAX_CALLSNoAdapter-attempt budget16
SEARCH_CACHE_MAX_ENTRIESNoMaximum number of persisted exact-result cache entries
SEARCH_BUDGET_MAX_RESULTSNoAdmitted raw-result budget100
HTTP_ALLOW_UNAUTHENTICATEDNoSet to true to allow unauthenticated HTTP mode
SEARCH_BUDGET_MAX_ELAPSED_MSNoEnd-to-end elapsed-time budget in milliseconds30000

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Server capabilities have not been inspected yet.

Tools

Functions exposed to the LLM to take actions

NameDescription

No tools

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Maintenance

ActivityActive
ResponsivenessResponsive