Skip to main content
Glama
Sign In
AI Guardian

AI Guardian

by AIops-tools
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

Related Servers

Alternatives to AI Guardian

No user-submitted related servers found.

    Related Servers

    • local-llm-mcp

      primal79
      A
      license
      A
      quality
      C
      maintenance
      Enables a cloud orchestrator to query a local Ollama model over Tailscale or a LAN as a side-channel for drafts, summaries, and other helper output, with tools to list available models and to redact text. Answers leaving the server are passed through output-only DLP that masks card numbers, IBANs, emails, phone numbers, API keys, and similar secrets before they reach the cloud.
      3
      MIT

    • mcp-gateway

      darrshangovender
      A
      license
      Not graded
      quality
      B
      maintenance
      Enables secure exposure of internal tools to LLM clients with API-key authentication, tenant isolation, per-tool guardrails, PII scrubbing, rate limiting, audit logging, and telemetry over stdio and HTTP/SSE.
      MIT

    • Ollama MCP Server

      wisniewski880
      A
      license
      Not graded
      quality
      A
      maintenance
      Enables MCP-compatible clients to access local or cloud LLMs, manage and run models, generate completions and embeddings, chat with tool/function support, and use web search/fetch through Ollama Cloud.
      923 npm
      AGPL 3.0

    • claude-token-saver-mcp

      BlackFoil
      A
      license
      A
      quality
      D
      maintenance
      Enables Claude Code to offload routine code generation and text processing tasks to a local Ollama LLM, saving Cloud API tokens and costs with automatic model selection and security features.
      11
      867 npm
      3
      Apache 2.0
    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A3.6/5.0

    Scored across 21 tools

    Disambiguation4/5

    Most tools target clearly distinct actions/resources: scanning (scan_prompt), generation (guarded_generate), chat (observe_chat), policy (allowlist/denylist), and state reads (server_status, policy_view). The one soft cluster is model introspection—list_models, running_models, model_details, and vram_usage—but their descriptions clearly separate installed, loaded, single-model, and aggregate-VRAM views.

    Naming Consistency3/5

    There is a loose but not reliable pattern: action tools are verb_noun (pull_model, remove_model, unload_model, scan_prompt, set_model_allowlist, pin_model_digest, undo_apply) while many state tools are noun phrases (server_status, policy_view, vram_usage, usage_events, anomaly_report, model_provenance). The mix is readable and grep-able, but not predictable enough to be a clean convention.

    Tool Count3/5

    At 21 tools this is on the heavy side for the rubric's 3-15 sweet spot, spanning model introspection, model lifecycle, policy/provenance, scanning/governance, usage analytics, undo, and proxy guidance. Most tools appear justified by a genuinely broad domain, but the surface is large enough that count is borderline rather than well-scoped.

    Completeness4/5

    Coverage is strong: model lifecycle (list/pull/remove/unload/details), policy management (view/allow/deny/pin/provenance), governance (scan, guarded_generate, observe_chat, usage_events, anomaly_report), and an undo mechanism (undo_list/undo_apply). Minor gaps remain—allowlist/denylist can only be replaced wholesale (no per-model toggle), and there is no persistent block-threshold configuration tool—but core workflows are covered without dead ends.

    Maintenance

    ActivityNo data
    ResponsivenessOutdated data

    Resources