Get a PR review
Fetches a PR review with its findings. Results are limited to repositories within the token’s rbac scopes.
Authorizations
API token obtained from the Tokens endpoint or CLI device login. Include as Authorization: Bearer <token>. Requests made with a managed CLI session also include X-Strix-Workspace: <organization_id> to pin a process to the workspace it started in; recovery endpoints report the current workspace after a concurrent switch.
Path Parameters
PR review id.
Response
PR review details.
One PR review run plus PR-wide reconciled finding counts in findings.
github, gitlab, bitbucket, azure_repos pending, running, completed, posting, posted, failed Source branch reviewed by this run.
Destination branch for the pull request. Null when the provider did not supply it.
webhook, manual review, verify_base, null The pull request's own lifecycle, stamped when a review starts and kept live by provider webhooks. Null on older reviews.
open, merged, closed, null approve, comment, request_changes, null Trigger-maintained count for this individual review run. Prefer the nested findings.total field for the PR-wide aggregate shown by list and detail views.
x >= 0Trigger-maintained unresolved count for this individual review run. Prefer nested findings.unresolved.total for the PR-wide actionable aggregate.
x >= 0PR-wide finding counts used by list and detail views. They combine the selected/latest run's reconciled findings with deduplicated fixed findings from earlier runs of the same pull request. The top-level severity totals cover that aggregate; unresolved covers findings still open or in progress, snoozed counts deliberately deferred findings, and fixed includes fixed history across runs.
Reason code behind the current status. Set when the launch path or the infrastructure set the status, or when a completed scan ended early near its cost limit (budget_near_limit). null for an ordinary engine-owned status.
sandbox_pending, config_missing, llm_not_configured, secrets_missing, connector_unavailable, credentials_unreadable, image_pull_failed, pod_unschedulable, pod_rejected, container_config_failed, volume_failed, sandbox_ended, sandbox_timeout, docker_unavailable, sandbox_create_failed, launch_failed, retest_already_running, watchdog_timeout, launch_no_telemetry, relaunch_failed, relaunches_exhausted, engine_failed, budget_near_limit Secret-free explanation of status_detail_code. Self-hosted installs return the exact infrastructure message, for example the Kubernetes scheduler message. The hosted product returns generic wording. For budget_near_limit every install returns the cost figures, for example The run used $96.13 of its $100.00 cost limit (96%)….
What a running review is doing right now: Reading the changes, Investigating or Verifying. Reported every 30 seconds while the review runs and kept after it finishes. null when no status was reported.
One line on what the review is doing within live_stage, for example Tracing how partial refund amounts are validated.
When live_stage and live_status were last reported.
GitHub login of the comment author who requested this review; null for automatic, manual and older reviews.
What started a webhook review: the PR was opened, a new commit was pushed (pr_synchronized), or Strix was mentioned. null for reviews started from the dashboard or the API.
pr_opened, pr_synchronized, mention, null One or two sentences from the reviewing agent on what the pull request changes. Set when the review finishes.
Suspected issues the review checked and rejected, each with the reason it is not a vulnerability. Set when the review finishes.