CSM Copilot
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_accountsA | Find accounts strictly past the contact threshold; rank by planned launch date. Defaults to a fixed synthetic snapshot. A generated draft is never a contact. Missing contact history is explicitly unknown. Live accounts are excluded by default. A later as_of recalculates elapsed time from the same dataset; no records are refreshed. Omit as_of or pass null for the snapshot. Empty date strings are invalid. |
| get_account_contextB | Read app/platform integrations, issues and attributed synthetic evidence for an account. |
| check_readinessA | Check six scoped requirements per app/platform, preserving unknown vs not_met. checklist_complete means only this project's checklist; it is not launch approval. |
| triage_issueA | Retrieve possible checks for blank Offerwall or reward callbacks. issue_text must contain customer prose, not an issue ID. For a recorded issue, retrieve its message with get_account_context or use draft_ticket(issue_id). Lexical baseline, not semantic diagnosis; negation/paraphrases can fail. Returns candidate categories, questions and sources, never a confirmed root cause. |
| draft_ticketB | Prepare a technical handoff from a recorded issue; expose absent fields. Does not submit. |
| draft_followupB | Prepare an evidence-bound bilingual template. Does not send or update contact history. A host may rewrite the wording but must preserve uncertainties and source references. |
| validate_draftB | Lint narrow promise patterns. No flags does not mean factual or semantic verification. |
| friction_digestA | Aggregate recurring checklist gaps and failures into Product review questions. Fixed snapshot; counts distinct accounts and integrations separately. Not a trend, defect diagnosis or model-generated clustering. Does not log or submit feedback. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| product_feedback_review | Turn recurring checklist friction into a reviewable Product discussion. |
| morning_review | A complete customer-success morning workflow for a model-enabled MCP host. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 8 tools
Each tool targets a distinct stage of the CSM workflow (list, context, readiness, triage, draft, validate, digest). The only mild overlap is between draft_ticket and draft_followup, but their descriptions clearly separate technical handoff from bilingual follow-up templates.
Seven of eight tools follow a clean verb_noun pattern (list_accounts, get_account_context, check_readiness, triage_issue, draft_ticket, draft_followup, validate_draft). friction_digest breaks the pattern as a noun_noun name, a minor deviation but still readable.
Eight tools is well-scoped for a customer-success copilot, covering listing, diagnosis, drafting, validation and aggregation without redundancy. Each tool earns its place in a coherent pipeline.
The surface covers the read/diagnose/draft/validate lifecycle for account management well. Gaps are intentional (no submit/send, no contact-history update), but there is no way to mutate records, which could limit some workflows.