impactdotcom-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| IMPACT_BASE_URL | No | API base URL override | https://api.impact.com |
| IMPACT_AUTH_TOKEN | Yes | Basic-auth password (secret) | |
| IMPACT_ACCOUNT_SID | Yes | Basic-auth username (your Account SID, starts with IR) | |
| IMPACT_ACCOUNT_TYPE | No | API persona: Mediapartners (publisher) or Advertisers (brand) | Mediapartners |
| IMPACT_ENABLE_WRITES | No | set true to register the write tools | false |
| IMPACT_MAX_CONCURRENCY | No | max concurrent API requests (rate-limit guardrail, 1–16) | 4 |
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 |
|---|---|
| diagnose_trackingA | PRIMARY diagnostic. For a date window, pulls the click → action → revenue funnel and returns a plain-English read of why you might see clicks but no revenue: whether conversions are reaching Impact at all, and in what state (PENDING / APPROVED / REVERSED, incl. TEST_ACTION). Optionally narrow by campaign/SubId or trace a specific test OrderId. Max 45-day range. |
| list_actionsA | List Actions (conversions/commissions). With no |
| get_actionA | Get a single Action by its Id, with all fields (State, Oid, SubId, Payout, EventDate, LockingDate). |
| find_action_by_orderA | Trace a specific order (e.g. a test order) by its OrderId via the Advanced Action Listing report, including the action's Status and reversal reason (Status Detail / DispositionCode). The plain Actions endpoint can't filter by OrderId. |
| get_clickA | Get a single click by its id (irclickid / im_ref) — landing page, referring URL, campaign, device, SubId. Use to confirm a specific click was recorded. |
| export_clicksA | List clicks for a single day (optionally filtered by program/SubId) via Impact's async ClickExport job. Returns click rows once the export completes. |
| list_action_inquiriesA | List action inquiries (disputes for missing/incorrect/rejected actions) you've submitted. |
| get_action_inquiryA | Get a single action inquiry by its Id, including its status and resolution. |
| list_reportsA | List the reports this Impact.com account can run. Only reports with ApiAccessible=true can be run via run_report. |
| get_report_metadataA | Get a report's available filters and columns. Call this before run_report to learn the parameter names. |
| run_reportA | Run a report synchronously and return its rows. Params vary by report (commonly StartDate, EndDate, ProgramId/SUBAID) — use get_report_metadata to discover them. Dates are YYYY-MM-DD. |
| run_report_exportA | Start an asynchronous report export (CSV/JSON) and poll until the result is ready. Use for large reports. Returns the result rows, or a job URI if it is still running at timeout. |
| list_campaignsA | List the programs/campaigns this partner has joined. Optionally filter by status (Active or Expired). |
| get_campaignB | Get details for a single program/campaign by its CampaignId. |
| list_catalogsA | List product catalogs available to this partner. |
| list_catalog_itemsB | List items in a product catalog by CatalogId. |
| get_accountA | Get the account summary for the configured AccountSID — name, id, and the list of REST APIs this token can access. Smallest call; useful to confirm auth, the persona prefix, and which scoped APIs are available. |
| impact_api_getA | Fallback read tool: GET any persona-scoped Impact endpoint not covered by a curated tool. Prefer curated tools first. Read-only. |
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 18 tools
Most tools are clearly separated by resource and action (list vs get). The fallback impact_api_get is intentionally generic but well described as a last resort; run_report and run_report_export could be confused but their sync/async distinction is clear.
The vast majority follow a consistent verb_noun pattern (list_*, get_*). The fallback impact_api_get deviates slightly from the pattern, but it's a unique name that's still readable.
At 18 tools, the server is on the heavier side but still well-scoped for the Impact.com API's breadth. Each tool addresses a distinct resource or mode (sync vs async, individual vs list).
The core read-only surface is well covered: campaigns, catalogs, actions, clicks, action inquiries, and reports. Missing write operations (e.g., create action inquiry) are a minor gap, partially mitigated by the fallback read tool, but the server appears focused on diagnostics and reporting.