{"id":"ta-lib-mcp-server","name":"ta-lib-mcp-server","homepage":"https://pypi.org/project/ta-lib-mcp-server/","repo_url":"https://github.com/yalcin/ta-lib-mcp","category":"ai-ml","subcategories":[],"tags":["mcp","ta-lib","technical-analysis","indicators","python","tooling","read-only","agents"],"what_it_does":"A read-only Python MCP server that exposes TA-Lib functionality via MCP tools for listing available indicators/categories, fetching indicator metadata, computing indicator values from numeric OHLCV-like inputs, and reporting version information. If TA-Lib is not installed, it still starts and returns actionable errors for TA-Lib-related tools.","use_cases":["Connect an MCP-capable agent (e.g., Claude Desktop, Codex, Gemini CLI) to TA-Lib indicator discovery and computation","Generate indicator metadata (inputs/outputs/parameters) for UI or agent reasoning","Compute technical-analysis indicators from provided time-series arrays without writing direct TA-Lib integration code","Build retrieval/selection workflows where an agent lists indicators and then computes specific ones"],"not_for":["Production trading execution systems requiring low-latency streaming or stateful order management","Scenarios requiring strict compliance/validation beyond local tool execution","Environments where arbitrary user-provided numeric inputs are not validated/sandboxed at the application layer"],"best_when":"You want an agent to programmatically browse and compute TA-Lib indicators in a local/offline, read-only manner using MCP tools.","avoid_when":"You need hosted APIs with SLAs, authentication for multi-tenant access, or you require strong operational guarantees not addressed by the provided documentation.","alternatives":["Direct TA-Lib Python integration in your app (Python + TA-Lib bindings)","A different TA-Lib wrapper exposing REST/GraphQL APIs","An MCP server or tool layer that wraps TA-Lib via a service you control with stricter API gateways and observability"],"af_score":65.5,"security_score":32.8,"reliability_score":23.8,"package_type":"mcp_server","discovery_source":["pypi"],"priority":"low","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-04-04T21:42:54.882094+00:00","interface":{"has_rest_api":false,"has_graphql":false,"has_grpc":false,"has_mcp_server":true,"mcp_server_url":null,"has_sdk":false,"sdk_languages":[],"openapi_spec_url":null,"webhooks":false},"auth":{"methods":[],"oauth":false,"scopes":false,"notes":"No authentication mechanism is described. For MCP, access is typically local/host-controlled; this package documentation does not mention auth/authorization controls."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"Open-source package; no pricing model indicated."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":65.5,"security_score":32.8,"reliability_score":23.8,"mcp_server_quality":78.0,"documentation_accuracy":70.0,"error_message_quality":null,"error_message_notes":null,"auth_complexity":90.0,"rate_limit_clarity":5.0,"tls_enforcement":50.0,"auth_strength":10.0,"scope_granularity":0.0,"dependency_hygiene":55.0,"secret_handling":60.0,"security_notes":"No auth is described (score reduced). The README indicates input validation via validators and a read-only design with no stateful/exchange operations, which is positive. No details are provided about transport (TLS) or how the MCP server is bound/listened, and no rate limiting or authorization controls are documented. Dependency hygiene is unknown from the provided data; a baseline score is used.","uptime_documented":0.0,"version_stability":30.0,"breaking_changes_history":20.0,"error_recovery":45.0,"idempotency_support":"true","idempotency_notes":"Most tool operations (listing, fetching info, computing from provided arrays) are naturally idempotent given the same inputs; no side effects/state are described.","pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["The server requires Python 3.13+ and MCP CLI support; agents may need a compatible runtime environment.","If TA-Lib is not installed, TA-Lib tools should fail with actionable errors; the agent should handle this fallback path.","Tool inputs are validated (per README), so malformed arrays/parameter types should produce validation errors that the agent must correct rather than retry blindly."]}}