Review genuinely new staging errors

Compare a bounded staging case window with prior fingerprints, inspect only genuinely new failure modes, and prepare a human-reviewed oncall handoff.

Updated · published

markdown for agents →

difficulty intermediate · time to value 5 minutes · execution human review required

Start from this

Review staging errors first seen in the last hour and compare them with the prior day. For each genuinely new fingerprint, summarize impact, suspect deploy, and next investigation. Do not message oncall or mutate cases.

Why this matters

Volume thresholds create noise in staging. The higher-value question is whether a fingerprint is genuinely new, reappeared after resolution, or began immediately after a deploy. That determination needs bounded case and deploy evidence, not an unrestricted agent with a webhook.

Run the review

Fetch the filtered OpenAPI schema for the cases and deploy reads, then use only their documented query parameters:

GET /v1/projects/:project_id/cases
GET /v1/cases/:case_id
GET /v1/projects/:project_id/deploys

Within the interactive session:

  1. Select staging cases first seen in the review window.
  2. Compare each fingerprint with the prior-day case window.
  3. Fetch details only for genuinely new or reappearing fingerprints.
  4. Correlate last_deploy_sha and recent deploy evidence.
  5. Separate proven facts from a suspected code cause.

Raw case details may contain user and request data. Minimize what enters the summary and do not copy secrets, tokens, or unnecessary identifiers.

Output format

New staging fingerprints — <UTC window>

<fingerprint>
- first seen / last seen / event count / affected users
- suspect deploy and evidence strength
- top in-app frame and likely owner
- recommended next investigation
- missing evidence

Return the handoff to the reviewer. A human decides whether to notify oncall, suppress noise, change case state, or start a code-fix workflow.

Instrumentation

Trusted server errors use the server-ingest credential:

await fetch("https://api.agentry.sh/v1/logs/", {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${process.env.AGENTRY_SERVER_API_KEY}`,
    "Content-Type": "application/json",
    "User-Agent": "myapp/1.0",
  },
  body: JSON.stringify({
    message: err.message,
    stack: err.stack,
    environment: "staging",
    release: process.env.APP_RELEASE,
  }),
});

Unattended boundary

There is no versioned staging-notification template. A runner cannot receive an owner key, raw case browsing, generic queries, or messaging authority. If a reviewed case should enter the supported code-fix path, use the exact [email protected] template and its policy gates; it may create a draft PR but still cannot message, merge, deploy, or declare production verification without the required trusted evidence.

Variations

  • “Review only fingerprints reappearing after a resolved state.”
  • “Compare the hour after a selected deploy with the hour before it.”
  • “Limit the review to fingerprints affecting at least five distinct users.”
  • “Draft an oncall handoff for a human to review; do not send it.”

Adapt this workflow in your own agent.

Ask your agent to map the starter prompt to your saved signal map, live events, and the current HTTP API before answering.

Agentry gives coding agents production context from errors, product analytics, and deploys through one HTTP API. Fetch and read https://agentry.sh/install.md completely, then install Agentry's full first-time telemetry baseline in this repository. I authorize device-code login, saving the owner key only to ~/.agentry/credentials.json, creating one Agentry project for this repository, and read-only inspection of shipped source plus hidden CI/provider configuration. Deeply analyze the product's business logic and complete value flow before choosing telemetry. Save the complete source-backed plan and show me its exact contents and hash. Until I explicitly approve that plan or request a replacement, do not edit app/CI source, place runtime/CI credentials, commit, exercise product traffic, start proof, push, or deploy. After approval, I authorize only the reviewed targets: place the required scoped browser/server/CI credentials through the established environment or secret mechanism, preserve existing telemetry, implement and test the baseline, commit it, push that reviewed commit when the shipped CI/provider path requires it, exercise safe proof paths with test/non-customer data, and perform one deployment through the reviewed shipped CI/provider path. Ask first if proof would charge money, contact a third party, change real customer data, or require new external access. After the plan is saved, immediately before every onboarding state-changing POST, GET current onboarding state, perform only its single returned next_action, then read state again; do not batch or infer later stages. Continue until status is verified, installation_complete is true, and next_action is null. Keep all secrets, source snapshots, proof markers, and scratch files outside the repository.

+ Full access
5.5 Extra High
  1. 1. Open your repo in Codex, Claude Code, Cursor etc.
  2. 2. Paste the install prompt.
  3. 3. Your agent reads the install doc and shows you an implementation plan for approval.