Fast.io

AI Chat Management

ai_manage

Destructive

Fastio AI chat (write): ask a question about workspace or share files and get a cited answer, a workspace status summary, create/update/publish/cancel chats, send messages, generate an AI share. Destructive: chat-delete. ask, status, chat-create and message-send consume AI credits. Reads are on ai. action='describe' gives the full action/param reference.

Input Schema

TableJSON Schema

Name	Required	Description
`kind`	No	chat-create (workspace only): 'user' (default) or 'agent' (agentic; immutable).
`name`	No	New chat name.
`files`	No	File opaque IDs (max 25, share share-generate).
`action`	Yes	Operation. Use 'describe' for full action reference.
`effort`	No	Reply effort for THIS message: 'normal' (default) or 'high' (deeper reasoning, no per-reply step limit; slower, more credits). Per message — resend it on each message that should use it.
`chat_id`	No	AI chat ID.
`privacy`	No	Chat privacy (default: private). Workspace chat-create only — share chats are always private.
`node_ids`	No	File node IDs (max 25, workspace share-generate).
`share_id`	No	profile_id alias; implies profile_type=share.
`context_id`	No	Alias for profile_id (either name works)
`profile_id`	No	19-digit workspace or share ID (pair with profile_type). Aliases: workspace_id / share_id (imply the type), context_id, instance_id.
`query_text`	No	Question or prompt (max 32,000 chars).
`instance_id`	No	Alias for profile_id.
`context_type`	No	Alias for profile_type (either name works)
`files_attach`	No	Files for the AI to read: comma-separated nodeId:versionId (versionId optional = current). Works without Deep Indexing; an unreadable node errors.
`profile_type`	No	Profile type: "workspace" or "share".
`workspace_id`	No	profile_id alias; implies profile_type=workspace.
`subject_files`	No	Files pinned as the turn's focus subjects (sent as `subjects`): comma-separated nodeId:versionId, same format and limits as files_attach.
`folders_attach`	No	Folders for the AI to read: comma-separated folder nodeIds (no depth). Denied on restricted-view shares.
`subject_folders`	No	Folders pinned as the turn's focus subjects (sent as `subjects`): comma-separated folder nodeIds, same format as folders_attach.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 17, 2026

addedInput schema / properties / effort

Added value: +{
+  "description": "Reply effort for THIS message: 'normal' (default) or 'high' (deeper reasoning, no per-reply step limit; slower, more credits). Per message — resend it on each message that should use it.",
+  "enum": [
+    "normal",
+    "high"
+  ],
+  "type": "string"
+}

Changed2 schema fields changedOct 10, 2026

addedInput schema / properties / subject_files

Added value: +{
+  "description": "Files pinned as the turn's focus subjects (sent as `subjects`): comma-separated nodeId:versionId, same format and limits as files_attach.",
+  "type": "string"
+}

addedInput schema / properties / subject_folders

Added value: +{
+  "description": "Folders pinned as the turn's focus subjects (sent as `subjects`): comma-separated folder nodeIds, same format as folders_attach.",
+  "type": "string"
+}

Changed1 schema field changedOct 7, 2026

changedInput schema / properties / files_attach / description

Previous value: -"Files for the AI to read: comma-separated nodeId:versionId (versionId optional = current). Works without intelligence; an unreadable node errors."New value: +"Files for the AI to read: comma-separated nodeId:versionId (versionId optional = current). Works without Deep Indexing; an unreadable node errors."

AddedOct 3, 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, readOnlyHint=false, and openWorldHint=true, so the safety profile is covered. The description adds value beyond that by identifying exactly which action is destructive (chat-delete) and by disclosing that ask, status, chat-create, and message-send consume AI credits. It stops short of stating reversibility or permission requirements.

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?

Three dense sentences, front-loaded with the write-vs-read distinction and the action inventory, with no filler. Every clause carries operational information (destructive action, cost, fallback reference).

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 20-parameter multi-action tool with no output schema, the description covers scope, destructive surface, cost, and the read alternative, and points to action='describe' for the full reference. It only lightly gestures at return content ('a cited answer', 'a workspace status summary'), which is the remaining gap.

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% across all 20 parameters, so aliasing, enum meanings, and limits are already documented. The description adds only marginal parameter-level meaning, mainly tying credit consumption to specific actions; baseline 3 is appropriate.

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?

States a specific verb and resource ('Fastio AI chat (write)') and then enumerates the concrete action surface: ask, status, chat create/update/publish/cancel, message-send, share-generate. It also explicitly distinguishes itself from the read counterpart ('Reads are on `ai`'), so an agent can separate it from the sibling `ai` tool without opening the schema.

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?

Names the alternative for reads ('Reads are on `ai`') and flags the destructive action ('Destructive: chat-delete') plus the credit-consuming actions. It does not, however, explain when to pick one action over another (e.g. chat-publish vs share-generate), instead deferring to action='describe', which is a reasonable routing move but shifts the guidance out of the description.

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

All Fast.io 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.

Resources

Need Help?