Skip to main content
Glama
Atlasent

atlasent-mcp

Official
by Atlasent
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Hybrid

AtlaSent — Evaluate (Remote API)

atlasent_evaluate

Evaluate an action against your hosted policies to receive a decision (allow, deny, hold, escalate) and a permit token when allowed.

Instructions

Evaluate an action against your published AtlaSent policies. Returns allow/deny/hold/escalate with a permitToken on allow. Use this when ATLASENT_MODE=remote and you need to gate an action against your hosted policy engine.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
contextNoKey-value context matched against constraint rules (include target/resource info here).
explainNoWhen true, populates risk_envelope.factors with a per-factor score breakdown
actor_idYesThe actor performing the action (e.g. 'user:alice', 'service:deploy-bot').
target_idNoTarget resource the permit is bound to (e.g. service name, artifact id, tool-call target). Presenting a different target at verify yields a binding mismatch.
action_typeYesCanon-backed Action Type (for example 'production.deploy' or 'agent.tool.invoke').
execution_payload_hashNoHash of the exact executed payload (tool-call arguments / artifact digest). Bind it at evaluate via execution_payload_hash; presenting a different hash at verify yields PAYLOAD_MISMATCH.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed10 schema fields changedv2.12.2
    • removedInput schema / properties / action
      Removed value: -{
-  "description": "What action is being performed (e.g. 'production.deploy', 'records.delete').",
-  "maxLength": 256,
-  "minLength": 1,
-  "type": "string"
-}
    • addedInput schema / properties / action_type
      Added value: +{
+  "description": "Canon-backed Action Type (for example 'production.deploy' or 'agent.tool.invoke').",
+  "maxLength": 256,
+  "minLength": 1,
+  "type": "string"
+}
    • addedInput schema / properties / actor_id
      Added value: +{
+  "description": "The actor performing the action (e.g. 'user:alice', 'service:deploy-bot').",
+  "maxLength": 256,
+  "minLength": 1,
+  "type": "string"
+}
    • changedInput schema / properties / context / description
      Previous value: -"Key-value context matched against constraint rules."New value: +"Key-value context matched against constraint rules (include target/resource info here)."
    • addedInput schema / properties / execution_payload_hash
      Added value: +{
+  "description": "Hash of the exact executed payload (tool-call arguments / artifact digest). Bind it at evaluate via execution_payload_hash; presenting a different hash at verify yields PAYLOAD_MISMATCH.",
+  "maxLength": 256,
+  "minLength": 1,
+  "type": "string"
+}
    • removedInput schema / properties / org_id
      Removed value: -{
-  "description": "Organization ID that owns the policy.",
-  "maxLength": 256,
-  "minLength": 1,
-  "type": "string"
-}
    • removedInput schema / properties / resource
      Removed value: -{
-  "description": "The resource being acted on (e.g. 'env:prod', 'db:customers').",
-  "maxLength": 256,
-  "minLength": 1,
-  "type": "string"
-}
    • removedInput schema / properties / subject
      Removed value: -{
-  "description": "The actor performing the action (e.g. 'user:alice', 'service:deploy-bot').",
-  "maxLength": 256,
-  "minLength": 1,
-  "type": "string"
-}
    • addedInput schema / properties / target_id
      Added value: +{
+  "description": "Target resource the permit is bound to (e.g. service name, artifact id, tool-call target). Presenting a different target at verify yields a binding mismatch.",
+  "maxLength": 256,
+  "minLength": 1,
+  "type": "string"
+}
    • changedInput schema / required
      Previous value: -[
-  "subject",
-  "action",
-  "resource",
-  "org_id"
-]New value: +[
+  "actor_id",
+  "action_type"
+]
  2. First observedv2.11.0

TDQS

A3.8/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description discloses the decision values and permitToken, which is important because there is no output schema. However, beyond the basic annotations it does not reveal side effects (e.g., whether an audit record is created), expiration or verification expectations, or behavior for hold/escalate beyond the enum-like return.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two tight sentences lead with the core action and return type, then give the mode condition. No wasted words and the key information is front-loaded.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a no-output-schema tool with six parameters and many siblings, the description covers the essential outcome and the primary selection condition. It does not explain how permitToken should be used afterward or expand on hold/escalate semantics, but the rich schema descriptions compensate for most gaps.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, and the prose description adds no parameter-level meaning beyond what the schema already provides. The schema itself supplies useful semantics such as the context object and binding behavior of target_id and execution_payload_hash, so the baseline of 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the action ('Evaluate an action against your published AtlaSent policies') and names the output (allow/deny/hold/escalate with a permitToken). It distinguishes remote mode from any local evaluation path, though it does not explicitly name sibling tools like atlasent_evaluate_many or evaluate.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It gives an explicit condition ('Use this when ATLASENT_MODE=remote') and the intended purpose (gate an action against the hosted policy engine). It does not state when to prefer alternatives such as atlasent_evaluate_many or atlasent_verify_permit, so exclusions are missing.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools