GuardEntry MCP Server
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MCP_PORT | No | HTTP server port | 3001 |
| MCP_TOOLS | No | Comma-separated tool allowlist | (all tools) |
| GUARDENTRY_API_KEY | Yes | API key from GuardEntry dashboard | |
| GUARDENTRY_BASE_URL | No | Override for local/staging | https://app.guardentry.ai |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| guardentry_chatA | Send a natural language message to GuardEntry. Can query compliance data, create risks/controls/evidence, generate reports. Write actions are queued for dashboard approval. |
| guardentry_list_risksB | List the organization's risk register. Returns risks with severity scores, status, and descriptions. |
| guardentry_compliance_statusA | Get compliance readiness status for a specific framework or all frameworks. |
| guardentry_list_pendingA | List pending API actions awaiting dashboard approval. |
| guardentry_action_statusB | Check the status of a specific API action by ID. |
| guardentry_invoke_agentA | Invoke a GuardEntry agent by ID or type. The agent executes its configured skills (vulnerability scanning, compliance checks, risk analysis, etc.) and returns results. Write actions are queued for dashboard approval. |
| guardentry_invoke_skillA | Invoke a single GuardEntry skill directly (e.g., compliance_check, risk_analysis, scan_vulnerabilities). Read skills execute immediately; write skills are queued for approval. |
| guardentry_list_skillsA | List all available GuardEntry agent skills with descriptions, grouped by category (research, analysis, write, output). |
| guardentry_evaluate_actionA | Evaluate whether a proposed agent action is allowed, blocked, or requires approval based on the agent's policy. Returns a decision with reasoning. Use before any agent executes a sensitive task, tool call, or prompt. |
| guardentry_get_agent_policyA | Retrieve the effective policy for an agent. If no explicit policy exists, returns the inferred default policy that was auto-generated for this agent. |
| guardentry_confirm_policyA | Confirm an inferred agent policy, making it the active confirmed policy. Use after reviewing an auto-generated policy to promote it from 'inferred' to 'confirmed'. |
| guardentry_update_policyA | Update an agent policy's rules — blocked actions, allowed tools, require-approval patterns, risk tolerance, and mode. A version snapshot is saved before each update. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 12 tools
Most tools target distinct resources and actions, such as listing risks, invoking agents/skills, and managing policies. The main potential confusion is between guardentry_invoke_agent and guardentry_invoke_skill, but their descriptions clarify that one invokes an entire agent while the other invokes a single skill. guardentry_chat is a broad catch-all that overlaps with some read operations, but its natural language interface makes it distinct enough.
All tool names share the guardentry_ prefix and use snake_case, which is good. However, the naming pattern is inconsistent: several tools follow a clear verb_noun structure (list_risks, invoke_agent, evaluate_action, get_agent_policy), while others lack a verb or use noun-only names (chat, compliance_status, action_status). This mixed convention makes the set slightly less predictable.
With 12 tools, the set is well-scoped for a governance/compliance platform covering risks, compliance, agents, skills, and policies. Each tool addresses a distinct concern, and the count stays within the ideal 3-15 range without feeling bloated or sparse.
The toolset covers core read operations and several actions, such as listing risks, invoking agents/skills, and updating policies. However, there are notable gaps: no dedicated create/update/delete tools for risks or controls, no tool to list agents, and no way to approve or reject pending actions directly. guardentry_chat can handle some writes indirectly, but this creates a dependency on a general-purpose tool.