Askli Customer Research
Server Details
Run interviews with real customers and retrieve evidence-backed product findings.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- OAuth
- Works in Glama
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 10 tools
Each tool targets a fairly distinct action: creating studies, reading plans/findings/status/transcripts, listing, and updating/confirming plans. The main overlap risk is among the plan-lifecycle tools (get_plan, update_plan, confirm_plan) and the read tools (get_findings vs get_plan vs get_study_status), but the descriptions clearly separate them by artifact and workflow stage.
All tools use snake_case with a leading verb (create_study, get_plan, update_plan, list_studies, confirm_plan, etc.), and even the noun phrases are consistent. There are no deviating conventions or camelCase mix-ins.
Ten tools is well within the ideal range and each maps to a meaningful step in the study lifecycle. No tool appears redundant or padded.
The surface covers the full flow from create_study through clarification, plan review/update/confirm, status checks, findings, transcripts, listing, and force-analysis. Minor gaps exist, such as no delete/cancel-study operation, but the core evidence-gathering lifecycle is fully covered.
Available Tools
10 toolsanalyze_nowAnalyze nowAInspect
Analyze finished interviews now instead of waiting for Askli's next checkpoint. Only run this if the user asks. Takes a few minutes. When presenting the result, keep Askli's caveats about small samples.
Inputs: studyId (required).
answer_clarificationAnswer clarifying questionAInspect
Send the user's answer to Askli's clarifying question. Present each question exactly as Askli asks it, and never answer on the user's behalf. Omit answer to re-fetch the current question. Returns the next question or says the plan is ready.
Inputs: studyId (required), answer.
confirm_planConfirm research planAIdempotentInspect
Confirm a reviewed plan so participants can be chosen. Requires user_approved: true, which you may only set after the user has seen the plan and explicitly approved it. Confirming doesn't charge anything.
Inputs: studyId (required), user_approved (required).
create_studyCreate studyAInspect
Start customer research with real human participants when a product decision needs evidence rather than synthetic model feedback. Pass the user's question in their own words; don't rephrase or improve it. Returns Askli's first clarifying question, or says the plan is ready. Out-of-scope questions create no study.
Inputs: question (required).
get_findingsGet findingsARead-onlyInspect
Read a study's Decision Review. Present it faithfully: keep Askli's counts (like "observed in 2 of 3 conversations") and caveats, and don't add conclusions or raise confidence. If there isn't enough evidence yet, say so plainly.
Inputs: studyId (required).
get_planGet research planARead-onlyInspect
Read a study's research plan. Returns structured data plus a text version in a fixed order (question, who, method, participants, themes, what it can and can't tell you, status with a link to the full plan); present that text as given.
Inputs: studyId (required).
get_study_statusGet study statusARead-onlyInspect
A study's stage, interview progress, payment status, and next step. Check this when the user returns from the payment page; it's the only way to know whether payment went through.
Inputs: studyId (required).
get_transcriptGet transcript quotesARead-onlyInspect
Participant quotes relevant to one finding (its ID or title from get_findings) or to a topic, each with the question it answered and a link to the full transcript on the website. For an authenticated study owner, Bring Your Own display names are shown when assigned; canonical participant IDs remain included. Never returns emails or unverified identity claims.
Inputs: studyId (required), topic, finding, participant.
list_studiesList studiesARead-onlyInspect
List the user's Askli studies, newest first, with each study's ID, question, and current stage.
Inputs: no named parameters.
update_planUpdate research planADestructiveIdempotentInspect
Change an unconfirmed plan's investigate, audience, screeningCriteria, or themes. Only apply changes the user asked for or agreed to; send only those fields. Askli re-checks the plan and reports anything it saved differently.
Inputs: studyId (required), themes, audience, investigate, +1 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Run in-product voice interviews with AI agents and analyze source-linked evidence.
AI-moderated qualitative customer interviews, study authoring, participant links, and CSV exports.
Run user and customer interviews with real people from your AI agent, by MCP or REST API. Vetted, fraud-screened participants from a 4M+ consumer panel or your own list. AI moderation that ladders into why. Voice, video or chat in 80+ languages. Findings come back cited to quotes.
Turn a business question into a customer research report from simulated interviews.
Related MCP Servers
- AlicenseCqualityBmaintenanceRun real user interviews from AI agents and retrieve structured insights with themes and verbatim quotes.55MIT
- AlicenseNot gradedqualityDmaintenanceEnables semantic search and grounded answering over customer-research interviews, with every answer traceable to source quotes.MIT
- FlicenseNot gradedqualityBmaintenanceAnalyzes customer interview transcripts to surface recurring pain points and synthesize actionable roadmap themes through MCP, with deterministic, dependency-free execution.8-
- AlicenseAqualityCmaintenanceEnables local discovery of research topics, interviews users to uncover their unique perspective, and turns insights into content ideas and short-form Reel scripts.10MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.