End-to-end PR reviewer

The end-to-end PR reviewer is a DinoAI agent that acts like a thorough senior reviewer: it reads the PR diff itself with read_pull_request, checks the linked Linear issue, verifies test and documentation coverage, traces downstream impact through get_lineage and get_exposures, and posts a single structured verdict (APPROVE / REQUEST_CHANGES / REJECT) to Slack.

Prerequisites

Linear connected (the agent calls get_linear_issue to fetch linked tickets). Recommended but not strictly required: if the PR has no Linear link, the agent notes it under SCOPE and continues.
Slack connected (the agent posts reviews to #pr-reviews via post_slack_message).
Familiarity with programmable agents and running an agent with Bolt.

Estimated time: 15 minutes.

Steps

Create the agent

In the Agent app, open Agents, select New agent, and start from the dbt-model-reviewer template (or Start from scratch). Fill in the builder fields with the content below. See Build an agent in the UI for a tour of the builder.Name: pr-reviewer-e2eRole:

Senior Analytics Engineering Reviewer responsible for end-to-end PR
review: spec alignment, code quality, tests, docs, and downstream
impact.

Goal:

For the PR referenced in the trigger message (or, if none is named,
the most recently updated open PR from list_pull_requests):
(1) read the PR title, description, and full diff via
    read_pull_request,
(2) call get_linear_issue to fetch the linked ticket and verify the
    PR delivers what the ticket asked for,
(3) read every changed .sql and .yml file,
(4) check test coverage on new or modified models,
(5) check documentation completeness in schema YAML,
(6) assess breaking-change risk using get_lineage for downstream
    models and get_exposures for dashboards and apps consuming them.
Post a single structured review to the PR Slack thread under the
headings: SCOPE / CODE / TESTS / DOCS / IMPACT / VERDICT.

Backstory:

You are thorough but never noisy. You always cite the file and line
for any issue you flag. If the PR description has no Linear link, say
so in SCOPE and continue. You never approve a PR where TESTS or DOCS
is FAIL.

Model: leave Auto selected.Allowed tools (everything else is refused, see the Tools Reference):

list_pull_requests, read_pull_request
read_file, search_files_and_directories, ripgrep_search
get_linear_issue
get_lineage, get_exposures
run_sql_query
post_slack_message

Output: set the Slack channel to #pr-reviews (or your team’s channel).

read_pull_request means you do not have to collect and paste the changed-file list into the task: the agent resolves the PR and reads the diff itself. get_lineage and get_exposures (dbt Discovery tools) ground the IMPACT section in the real dependency graph, including dashboards, instead of a text search.

Deploy the agent

Select Deploy and choose where the agent lives: Just here saves it to the workspace and makes it live immediately, Open a pull request commits it to your repo as .dinoai/agents/pr-reviewer-e2e.yml so it is governed as code. Both appear in the Bolt agent picker. See Creating Agents in the App.

Run it on every PR with Bolt

Add a Run Paradime DinoAI Agent command to a Bolt schedule and trigger it on pull request merge, so the reviewer runs on every PR. The command runs natively with no API keys to configure. See Run an agent with Bolt for the full walkthrough.Pick pr-reviewer-e2e and set the Task:

Review the current PR end-to-end. Read the PR title, description, diff,
and any linked Linear issue. Check that the PR delivers what the ticket
asked for, that new or modified models have test coverage and
documentation, and assess downstream breaking-change risk via lineage
and exposures. Post one structured review to #pr-reviews under
SCOPE / CODE / TESTS / DOCS / IMPACT / VERDICT.

Read the verdict

After the agent runs, open #pr-reviews. The agent posts a single structured message with a clear verdict. It always cites the file and line number for every issue it raises, and it never returns APPROVE if either TESTS or DOCS is FAIL.

SCOPE   — does the PR match the Linear ticket? (PASS / PARTIAL / FAIL)
CODE    — correctness, naming, ref/source usage
TESTS   — coverage on new or changed models
DOCS    — descriptions on model + columns
IMPACT  — downstream models / exposures touched (via lineage)
VERDICT — APPROVE / REQUEST_CHANGES / REJECT

DinoAI PR reviewer posting a structured SCOPE / CODE / TESTS / DOCS / IMPACT / VERDICT review in the Slack channel

On a PR with full coverage that matches its Linear ticket, the review posts to #pr-reviews with an APPROVE verdict. On a PR missing tests or docs, TESTS or DOCS reads FAIL and the verdict is REQUEST_CHANGES or REJECT, with each issue cited by file and line. If the PR has no Linear link, SCOPE notes it and the rest of the review still runs.

How it works

The agent resolves the target PR (from the trigger message, or the most recently updated open PR), reads its full diff with read_pull_request, then works through the six review dimensions. For IMPACT it walks the dependency graph with get_lineage from each changed model and checks get_exposures for dashboards and apps that consume them, so “this change breaks the revenue dashboard” is a graph fact, not a guess.

Delegate test-writing (optional)

Instead of just flagging missing tests in the verdict, let the reviewer fix them: add test-maintainer to this agent’s Squad in the builder and invoke_agent to its allowed tools. The reviewer then delegates to the dbt™ test maintainer, which writes and validates the tests and reports back via notify_parent_session before the final verdict is composed.

Block CI on a REJECT verdict (optional)

Bolt’s on-merge trigger reviews PRs after they merge. If you want the review to gate the merge instead, trigger the agent from your CI with the API and fail the job on a REJECT:

run = paradime.dinoai_agents.trigger_run_and_wait(
    agent="pr-reviewer-e2e",
    message=f"Review PR #{pr_number} end-to-end and post the verdict to #pr-reviews.",
    timeout=1800,
)

verdict_line = next(
    (line for line in run.messages[-1].content.splitlines() if line.startswith("VERDICT:")),
    "VERDICT: UNKNOWN",
)
if "REJECT" in verdict_line:
    raise SystemExit("PR rejected by DinoAI reviewer")

This needs API credentials with DinoAI agent API capabilities stored as CI secrets. See the API and SDK reference.

Next steps

Run an agent with Bolt

Trigger this agent on a schedule, on merge, or in Turbo CI.

dbt™ test maintainer

Delegate test-writing to this sub-agent from the reviewer.

dbt™ documentation backfiller

Backfill missing descriptions on the same PR.

Programmable Agents reference

The agent schema, tools, and API.

Foundations

Automate with AI agents

Set up & configure

Orchestrate data pipelines

Migrate to Paradime

Develop in Paradime

Learn dbt™

End-to-end PR reviewer

Steps

How it works

Delegate test-writing (optional)

Block CI on a REJECT verdict (optional)

Next steps

Run an agent with Bolt

dbt™ test maintainer

dbt™ documentation backfiller

Programmable Agents reference

​Steps

​How it works

​Delegate test-writing (optional)

​Block CI on a REJECT verdict (optional)

​Next steps

Run an agent with Bolt

dbt™ test maintainer

dbt™ documentation backfiller

Programmable Agents reference

Steps

How it works

Delegate test-writing (optional)

Block CI on a REJECT verdict (optional)

Next steps