{"id":"shared-context-cache-mcp-server","name":"shared-context-cache-mcp-server","af_score":52.0,"security_score":36.0,"reliability_score":18.8,"what_it_does":"A Python MCP server that provides a shared cache for AI-agent computed results, including a trust/verification mechanism where agents can store results, confirm accuracy, and retrieve only entries that meet a configurable minimum trust threshold. Cache entries expire via TTL and basic/advanced analytics tools are exposed via MCP tools.","best_when":"Multiple cooperating agents (or teams) repeatedly compute the same deterministic-ish results and you want lightweight cross-agent sharing with a consensus-style confidence score.","avoid_when":"Confirmations and cache contents are influenced by untrusted or potentially malicious actors, or when you require cryptographic guarantees for provenance/authenticity.","last_evaluated":"2026-04-04T21:44:51.222586+00:00","has_mcp":true,"has_api":false,"auth_methods":[],"has_free_tier":false,"known_gotchas":["Trust score increases via confirmations; repeated confirmations by the same agent (or duplicate events) could inflate trust unless the server deduplicates (not documented).","get_trusted uses a configurable threshold; ensure consistent min_trust settings across agents for expected behavior.","Cache key conventions are important; inconsistent keying will reduce cache hit rate.","Trust persistence is local to the user (~/.shared_context_cache_trust.json); running multiple deployments/users may not share trust state as expected."],"error_quality":0.0}