replyatlas-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| REPLYATLAS_API_KEY | Yes | Your ReplyAtlas API key, e.g. mf_live_your_key_here. Create an API key in ReplyAtlas → Settings → API Keys. Requires a ReplyAtlas plan with API access. | |
| REPLYATLAS_BASE_URL | No | Override the ReplyAtlas API host. | https://replyatlas.com |
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 | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_accountA | Return the authenticated ReplyAtlas account: profile, plan tier, DM quota/usage, and feature flags. |
| list_ig_accountsA | List the connected Instagram accounts (id, handle, followers). Use an id as |
| list_automationsA | List comment→DM automations (paginated). Pass |
| get_automationC | Fetch a single automation by id. |
| create_automationB | Create a comment→DM automation. Required: name, igAccountId (from list_ig_accounts), triggerType, templateText. templateText supports {{handle}} and {{commentText}} tokens. Advanced/paid options are left at their server defaults. |
| update_automationB | Update fields on an existing automation. Only provided fields change. Set |
| delete_automationA | Permanently delete an automation by id. This cannot be undone. |
| list_leadsA | List/filter leads (paginated). Filters: minScore/maxScore (0–100), sentiment (POSITIVE|NEUTRAL|NEGATIVE), converted (true|false), q (search handle/name), tag (tag id). Page with |
| get_leadA | Fetch a single lead by id (includes activity). |
| list_tagsA | List the account's lead tags (id + name). Use a tag id with tag_lead or the list_leads |
| tag_leadA | Attach an existing tag to a lead. |
| export_leadsB | Export all leads as CSV text. |
| get_dashboardA | Dashboard snapshot (DMs sent, leads, conversions) for a range. range = today | 7d | 30d | cycle | custom. For custom, also pass from/to (ISO dates). |
| get_analyticsC | Analytics breakdowns. kind = overview | timeseries | sentiment | top-automations | by-trigger | follow-unlock. range = 7d | 30d | 90d. |
| list_dm_logsA | List sent-DM logs (paginated). Filters: automationId, status (comma-separated: QUEUED,SENT,DELIVERED,OPENED,CLICKED,FAILED), q (search). Page with |
| list_conversationsB | List human-agent inbox conversations (paginated). Filters: igAccountId, unreadOnly. Requires a plan that includes the human-agent inbox. |
| reply_to_conversationA | Send a DM reply in a conversation. SAFETY: call once with {id, text} to get a preview and a confirmToken; call again with the same id, text, AND confirmToken to actually send. Requires a plan that includes the human-agent inbox. |
| list_broadcastsA | List the account's broadcasts (newest first). Requires the Pro plan (BROADCAST). |
| create_broadcastA | Create a DRAFT broadcast (does NOT send). Provide exactly one audience: |
| send_broadcastA | Send a DRAFT broadcast to its audience. SAFETY: call once with {broadcastId} to preview the recipient counts and get a confirmToken; call again with the same broadcastId AND confirmToken to actually send. Requires the Pro plan. |
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 20 tools
Each tool maps to a distinct resource+action (automations, leads, tags, broadcasts, conversations, logs), and descriptions carefully clarify boundaries. The main risk is get_dashboard vs get_analytics, which both cover reporting and could be confused, and list_dm_logs vs list_conversations overlap slightly around messaging history.
All 20 tools follow a strict snake_case verb_noun pattern (list_*, get_*, create_*, update_*, delete_*, send_*, export_*, reply_to_*, tag_lead). Resource naming is predictable and the verb set is used consistently across resources.
20 tools is on the heavier side but justified by a broad SaaS domain spanning automations, leads, analytics, DM logs, inbox conversations, and broadcasts. Each tool maps to a real operation rather than padding, though it sits near the upper bound of reasonable.
Automations get full CRUD and leads/broadcasts/conversations cover core lifecycle flows. Minor gaps remain: no create_tag (tag_lead explicitly cannot create tags), no get_broadcast/update_broadcast/delete_broadcast, and leads are read-only, but these are workable for the primary workflows.