ShapelessAI

Create or edit an agent

agents_save

Destructive

Create an agent (no id) or edit one (with id). Setting autoPublishPosts arms publishing without human approval and needs the publish scope; everything else needs write.

Input Schema

Name	Required	Description
`id`	No	The agent id to edit; omit to create.
`name`	No	Name, up to 60 characters.
`prompt`	No	The standing brief, up to 4000 characters.
`status`	No
`planDays`	No	Wake days, 0=Sunday.
`timezone`	No	IANA timezone.
`planHours`	No	Wake hours.
`subreddits`	No	Up to 5 subreddits.
`connectionIds`	No	Accounts to post through; null means all.
`idempotencyKey`	No	On a create: Any string unique to this one intended thing (a UUID you keep). A retry with the same key answers what the first call made, with replayed: true, and makes nothing new. Derive it from the thing, not the attempt, so every agent and every restarted run that means the same one sends the same key.
`monthlyCapCents`	No	Monthly spend cap in cents; null = no cap. The studio shows it in credits.
`autoPublishPosts`	No	DANGEROUS: publish with no human approval. Needs the publish scope.

Output Schema

Name	Required	Description	Default
`strategy`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedNov 6, 2026

addedInput schema / properties / idempotencyKey

Added value: +{
+  "description": "On a create: Any string unique to this one intended thing (a UUID you keep). A retry with the same key answers what the first call made, with replayed: true, and makes nothing new. Derive it from the thing, not the attempt, so every agent and every restarted run that means the same one sends the same key.",
+  "type": "string"
+}

Changed1 schema field changedOct 7, 2026

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "additionalProperties": true,
+  "properties": {
+    "strategy": {
+      "additionalProperties": true,
+      "properties": {
+        "autoPublishPosts": {
+          "type": "boolean"
+        },
+        "id": {
+          "type": "string"
+        },
+        "monthlyCapCents": {
+          "type": [
+            "number",
+            "null"
+          ]
+        },
+        "name": {
+          "type": "string"
+        },
+        "planDays": {
+          "items": {
+            "type": "number"
+          },
+          "type": "array"
+        },
+        "planHours": {
+          "items": {
+            "type": "number"
+          },
+          "type": "array"
+        },
+        "prompt": {
+          "type": "string"
+        },
+        "spentThisMonthCents": {
+          "type": "number"
+        },
+        "status": {
+          "type": "string"
+        },
+        "timezone": {
+          "type": "string"
+        }
+      },
+      "required": [
+        "id"
+      ],
+      "type": "object"
+    }
+  },
+  "required": [
+    "strategy"
+  ],
+  "type": "object"
+}

First observedSep 29, 2026

TDQS

A4.2/5.0

Behavior4/5

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

Annotations already declare destructiveHint=true and openWorldHint=true, so the safety profile is partly covered; the description adds genuinely new context by disclosing the permission model (publish scope vs write) and flagging that autoPublishPosts removes human approval from publishing. It does not explain edit-overwrite semantics (whether omitted fields are cleared) or the replayed/idempotency behavior, which are the remaining gaps.

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 sentences, no filler, and the create-vs-edit rule is front-loaded ahead of the permission caveat. Every clause carries information an agent needs before calling.

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?

With an output schema present and annotations covering safety, the description only needs to cover mode selection and permission gating, which it does. For a 12-parameter mutation tool it is close to complete, though it says nothing about how edits treat unspecified fields or how autoPublishPosts interacts with connectionIds.

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 92%, so the schema already carries parameter meaning, including the autoPublishPosts danger note and the detailed idempotencyKey guidance. The description's mentions of id and autoPublishPosts largely restate what the schema says, adding scope requirements but little else beyond it. Baseline 3 applies.

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

Purpose5/5

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

The description names a specific verb and resource and splits the two operating modes by an observable signal: no id means create, an id means edit. That discriminator is stated up front, so an agent knows which behavior it will get before opening the schema, and no sibling (agents_list, agents_wake) offers competing create/edit semantics.

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 explicit routing logic between the two modes and states the prerequisite scopes (publish scope for autoPublishPosts, write for everything else). It stops short of naming alternatives or when not to use this tool, which keeps it out of the top band.

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

All ShapelessAI tools Inspect

Try in Browser

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

Glama MCP Gateway

Add one secure layer between your agents and this server.