{"id":"haiphucnguyen-askimo","name":"askimo","homepage":"https://askimo.chat","repo_url":"https://github.com/haiphucnguyen/askimo","category":"ai-ml","subcategories":[],"tags":["ai-ml","desktop-app","cli","rag","mcp","langchain4j","local-first","sqlite","automation","scripting","kotlin"],"what_it_does":"Askimo is a local-first AI desktop app and CLI that connects users to multiple LLM providers (cloud and local), supports persistent chat sessions stored locally (SQLite), performs document/code search with hybrid RAG (BM25 + vector), integrates MCP tools (via stdio or HTTP), and can run scripts (Python/Bash/JavaScript) from the chat. It claims local-only telemetry (token/cost/RAG performance) with data kept on disk.","use_cases":["Local-first RAG over personal folders/files and optionally web URLs","Interactive chat with persistent sessions stored locally","Connecting external toolsets via MCP servers to extend model capabilities","Running scripts from chat for automation (Python/Bash/JavaScript)","Using multiple LLM providers (including OpenAI-compatible endpoints) with per-session configuration","Vision-enabled multimodal conversations using supported multimodal models"],"not_for":["A hosted multi-tenant SaaS where you need centralized administration or server-side RBAC","Use cases requiring a documented public REST/SDK API for programmatic third-party integration","Security-sensitive environments that cannot tolerate executing model-driven scripts or tool calls without strict sandboxing"],"best_when":"You want a local desktop/CLI AI workspace that keeps conversation and indexed state on disk while integrating with MCP tools and local models (e.g., Ollama).","avoid_when":"You need a stable, documented network API surface (REST/OpenAPI/SDK) for external automation, or you cannot control/sandbox tool and script execution triggered by the agent.","alternatives":["LibreChat","Open WebUI (with Ollama/LM Studio)","Continue (editor-focused agent tooling)","LangChain-based local RAG apps","Anytype/Obsidian plugins for local RAG (varies by stack)","Other MCP clients (MCP-capable desktop clients)"],"af_score":39.5,"security_score":47.8,"reliability_score":31.2,"package_type":"mcp_server","discovery_source":["github"],"priority":"high","status":"evaluated","version_evaluated":null,"last_evaluated":"2026-03-30T15:19:05.330261+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":["Provider API keys for cloud models (as described: API key for cloud providers)","Local model configuration (e.g., point to a running Ollama instance / local endpoints)"],"oauth":false,"scopes":false,"notes":"No OAuth flow or scoped auth model is described in the README. Authentication appears to be per-provider (e.g., API keys) and connection configuration for local endpoints."},"pricing":{"model":null,"free_tier_exists":false,"free_tier_limits":null,"paid_tiers":[],"requires_credit_card":false,"estimated_workload_costs":null,"notes":"README does not describe pricing; as an open-source app, costs depend on the selected provider usage for cloud models."},"requirements":{"requires_signup":false,"requires_credit_card":false,"domain_verification":false,"data_residency":[],"compliance":[],"min_contract":null},"agent_readiness":{"af_score":39.5,"security_score":47.8,"reliability_score":31.2,"mcp_server_quality":50.0,"documentation_accuracy":55.0,"error_message_quality":0.0,"error_message_notes":null,"auth_complexity":80.0,"rate_limit_clarity":10.0,"tls_enforcement":70.0,"auth_strength":55.0,"scope_granularity":20.0,"dependency_hygiene":40.0,"secret_handling":50.0,"security_notes":"README emphasizes local-first operation and claims no data is uploaded (nothing uploaded for telemetry), which is positive. However, security-relevant details are not provided: TLS enforcement for any HTTP MCP connections or provider calls is not specified; auth details (scopes/least privilege) are not described beyond “API key for cloud models”; and script/MCP tool execution implies potential side effects unless the app implements strong sandboxing and prompt/tool safety controls (not described in the provided README).","uptime_documented":0.0,"version_stability":55.0,"breaking_changes_history":40.0,"error_recovery":30.0,"idempotency_support":"false","idempotency_notes":null,"pagination_style":"none","retry_guidance_documented":false,"known_agent_gotchas":["Tool execution risk: the README indicates it can run scripts and connect external tools via MCP; agents should assume tool execution side effects and plan for safety/sandboxing.","Local-only state: persistent sessions and indexes are local; automation agents may need to manage filesystem/workspace context for reproducibility.","Provider variability: multi-provider support can lead to different behaviors/limitations across providers and OpenAI-compatible endpoints."]}}