Research the agents already using your product
Start with one workflow where agents have real experience. Give them a direct way to report what was confusing, missing, broken, or great. Then turn repeated reports into a product decision.
Start from the surface agents already useEach route carries its own example report, workflow, boundary, and evidence rules.
By the surface agents use
The product is already in front of agent users. Start where the traffic is.
MCP server feedback
Let AI agents tell you which MCP tools were confusing, missing, or great. Collect direct feedback without wrapping or recording existing tool calls.
docs_feedback · search_repositories“The search worked, but I had to guess whether owner accepted a login or an organization name. One example would have removed two retries.”
Read the use case APIsAPI feedback from agents
Collect direct product feedback from the AI agents using your API. Find confusing schemas, missing endpoints, and successful but painful workflows.
schema_feedback · list_customers“The email filter works, but it is not in the schema. I only discovered it after reading a separate example.”
Read the use case Developer toolsDeveloper tool research
Research AI agents that use your CLI, SDK, or developer platform. Collect direct feedback about confusing commands, missing capabilities, and docs.
friction · deploy“The deploy completed, but the command returned before the preview URL was ready. I polled status six times without guidance.”
Read the use caseBy the question you need answered
The workflow is chosen. Now pick the research act that informs the decision.
Agent exit surveys
Ask an AI agent one short question after it completes a task. Collect direct answers inside the workflow without a separate survey page.
workflow_feedback · complete_checkout“The task succeeded. The hardest part was finding the customer identifier because I started with an email address.”
Read the use case Feature discoveryAgent feature requests
Collect feature requests from AI agents with the attempted job, workaround, and affected tool attached. Turn missing primitives into product decisions.
feature_request · create_ticket“I created 40 tickets one by one. A bulk endpoint would make this one operation and reduce partial failure handling.”
Read the use case DocumentationAgent docs feedback
Let AI agents report missing examples, ambiguous parameters, stale instructions, and recovery gaps in developer documentation.
docs_feedback · oauth_connect“The refresh flow is not shown in the quickstart. I inferred it from the token response and retried after the first expiry.”
Read the use case Workflow frictionSuccessful but painful work
Find AI agent workflows that technically succeed but require too many calls, retries, tokens, or hidden assumptions to finish safely.
workflow_feedback · checkout“Checkout took four sequential calls because customer lookup, address validation, quote, and confirmation cannot be composed safely.”
Read the use caseEvery route in this collectionThe same list a crawler sees, with the question each page answers.
One decision. One workflow. One real report.
Prove the research channel reaches an agent user before expanding exposure. A working endpoint is not the same as an active study, and filing is bimodal: where the tool and the filing norm are visible, agents file. Where they are hidden, nothing arrives.
- One paste to install
- 0ms on your hot path
- 14-day assisted pilot
- Two added tools to remove