Sunaiva Gate — Agent Rule Enforcement

Measured conclusion: Deep-tested. L2 outcome: tools_ok.

MCP enforcement layer that intercepts AI agent actions and blocks rule violations before execution.

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "Sunaiva Gate — Agent Rule Enforcement": {
      "url": "https://mcp.sunaivacore.io",
      "transport": "streamable-http"
    }
  }
}

Measured tools (16)

• validate_action

  Check if a proposed action passes all active validation rules. Returns allowed/blocked with rule violations.

  Parameters:

  • action (required) — string — The action to validate
  • context — string — Optional additional context, appended to `action` before rule matching — it can change which rules match and therefore the verdict.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• log_bypass

  Record an intentional rule bypass with justification. Cannot bypass constitutional rules.

  Parameters:

  • rule_id (required) — string — ID of the rule to bypass
  • reason (required) — string — Justification for the bypass

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_rules

  List active validation rules — the customer's TRUE enforced set (constitutional + recommended-default metadata on dev tier, all 69 premium rules on pro+). `category`/`preset` filter args are accepted but not applied server-side.

  Parameters:

  • preset — string — Filter by preset name
  • category — string — Filter by category

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• update_rules

  [Planned — not yet implemented on the remote endpoint, returns NOT_IMPLEMENTED] Enable/disable is not backed by any route in this API; use set_rule_mode (shadow/advisory/enforce) instead, or the local `npx @sunaiva/gate` package.

  Parameters:

  • enable — array — Rule IDs to enable
  • disable — array — Rule IDs to disable

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_audit_log

  View recent gate decisions — blocks, warnings, passes, and bypasses.

  Parameters:

  • limit — number — Max entries to return (default 50)
  • rule_id — string — Accepted for forward-compatibility but NOT applied server-side — the backend only honors limit/cursor. Filter the returned entries client-side.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• ship_confidence_check

  [Planned — not yet implemented on the remote endpoint, returns NOT_IMPLEMENTED] Ship Confidence Gate (Rule 42) dual-tier authorization is not yet built in this backend.

  Parameters:

  • artifact_id (required) — string — The artifact being shipped (e.g. '@sunaiva/[email protected]')
  • label — string — Human-readable label of the publish action (optional)
  • command_preview — string — First 300 chars of the command that triggered the check (optional)

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• audit_verify

  [Planned — not yet implemented on the remote endpoint, returns NOT_IMPLEMENTED] Forward-linked hash-chain / ed25519 audit-log verification is not yet built in this backend.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• set_rule_mode

  Set enforcement mode (shadow/advisory/enforce) for a specific rule. Requires API key (Pro+).

  Parameters:

  • rule_id (required) — string — Rule ID to configure
  • mode (required) — string — Enforcement mode
  • agent_id — string — Optional: scope override to a specific agent ID

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_rule_modes

  List all enforcement-mode overrides configured for the authenticated customer. Requires an account (Dev+).

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• reset_rule_mode

  Remove a mode override for a rule, reverting it to its default enforcement mode. Requires API key (Pro+).

  Parameters:

  • rule_id (required) — string — Rule ID whose override to remove

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• add_custom_rule

  Create a custom detection rule in the backend. Requires API key (Pro+).

  Parameters:

  • name (required) — string — Rule name
  • description (required) — string — Rule description (required — the backend 400s on a missing/empty description)
  • detection_pattern (required) — string — Compiled regex detection pattern (required — the backend 400s on a missing/empty detection_pattern). There is no 'pattern' field: 'detection_pattern' is the only field name the evaluator reads; the backend explicitly rejects 'pattern'.
  • severity (required) — string — Severity on match
  • tier — string — Optional tier restriction

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• update_custom_rule

  Update an existing custom rule. Requires API key (Pro+).

  Parameters:

  • id (required) — string — Rule ID to update
  • name — string
  • tier — string
  • severity — string
  • description — string
  • detection_pattern — string — There is no 'pattern' field — 'detection_pattern' is the only field name the evaluator reads; the backend explicitly rejects 'pattern'.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• delete_custom_rule

  Delete a custom rule. Requires API key (Pro+).

  Parameters:

  • id (required) — string — Rule ID to delete

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• list_custom_rules

  List all custom rules for the authenticated customer. Requires API key (Pro+).

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• list_presets

  List available rule preset bundles. Requires API key (Dev+).

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• apply_preset

  Apply a preset bundle, activating all its included rules. Requires API key (Pro+).

  Parameters:

  • preset_id (required) — string — ID of the preset to apply

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

Publisher website

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.