Skip to main content
Glama

preview_ai_dm

Try an AI DM brief before publishing it: what would the AI actually say to this follower message? Nothing is sent to Instagram and nothing is saved.

Pass the brief (ai_goal, ai_context, ai_exit_condition, ai_tone, ai_link — the same fields create_automation takes) and transcript, the conversation so far as [{role: "user"|"assistant", text}], ending with the follower's message to answer. To see a longer exchange, add the AI's reply as an "assistant" turn plus the follower's next message and call again. It runs the same model call the live automation does, so the answer is representative — but it varies from run to run.

Uses a Gemini call each time, so it is limited to 30 per 10 minutes and needs a plan with AI DM replies. respond=false means the inquiry filter would stay quiet on a first message like this one.

disclosure is the AI-assistant notice the follower gets as its own message just before the AI's first reply in a thread — "This is an automated reply from @handle's AI assistant. A real person sees your messages too." It is sent automatically, once per thread, and cannot be turned off, so show it to the user along with reply; it is empty when the transcript already holds an assistant turn.

The transcript is sent to Google Gemini to write the reply (see reeldrop.io/privacy).

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "workspace_id": "<workspace_id>",
      "ai_goal": "<ai_goal>",
      "transcript": "<transcript>"
    },
    "name": "preview_ai_dm"
  }
}

Input Schema

NameRequiredDescriptionDefault
ai_goalYes
ai_linkNo
ai_toneNofriendly
ai_contextNo
transcriptYes
workspace_idYes
ai_exit_conditionNo

Output Schema

NameRequiredDescriptionDefault
doneYes
replyYes
respondNo
disclosureNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.6/5.0
Behavior5/5

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

Goes well beyond the annotations: rate limit (30 per 10 minutes), plan requirement for AI DM replies, per-call Gemini cost, run-to-run variability, what `respond=false` means, and the exact disclosure message with its once-per-thread, non-optional semantics. The 'nothing is sent to Instagram and nothing is saved' claim is consistent in intent with destructiveHint=false; the tension with readOnlyHint=false is explainable by the external Gemini call and quota consumption, which openWorldHint=true already signals.

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

Conciseness4/5

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

Front-loaded with the one-line purpose and the 'nothing is sent/nothing is saved' reassurance before any detail. Five short paragraphs are mostly load-bearing, though the disclosure paragraph is long enough that its core takeaway ('always show it alongside reply; empty when an assistant turn exists') could be surfaced faster.

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

Completeness5/5

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

For a 7-parameter, externally-charged preview tool with an output schema already present, nothing an agent needs is missing: cost/quota, plan gating, transcript extension pattern, and the three return-relevant concepts (reply, respond, disclosure) are all covered. Return-value documentation is correctly left to the output schema.

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

Parameters4/5

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

With 0% schema description coverage the description must carry the parameter burden, and it largely does: it groups the ai_* brief fields as matching create_automation, and gives the transcript shape [{role: 'user'|'assistant', text}] ending with the follower's message. Gaps remain — workspace_id is never explained and ai_exit_condition, ai_tone defaults, and the 2000-char text cap are unaddressed.

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 — preview/try an AI DM brief and see what the AI would say to a follower message — and immediately distinguishes it from the live path by naming create_automation as the tool with the same fields. An agent can tell this apart from create_automation without opening either 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?

Gives a clear when-to-use context ('before publishing it') and a concrete continuation recipe (append the assistant reply plus the follower's next message and call again for a longer exchange). It stops short of an explicit exclusion such as 'do not use this to create the live automation' or naming a sibling to use instead, so it is strong but not exhaustive.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources