--- title: "Put an agent on call to post an evidence-grounded analysis on every CI and CD failure" description: "An agent joins the on-call channel, picks up each CI and CD failure, and posts an analysis grounded in the evidence it found for the person on call to review." url: https://starsling.dev/ci-playbooks/techniques/first-responder-agent canonicalUrl: https://starsling.dev/ci-playbooks/techniques/first-responder-agent --- # Put an agent on call to post an evidence-grounded analysis on every CI and CD failure An agent joins the on-call channel, picks up each CI and CD failure, and posts an analysis grounded in the evidence it found for the person on call to review. ## How to choose an approach This technique currently has one documented company example. Validate its prerequisites, tradeoffs and original CI environment against your repository before adapting it. - **[Anthropic](https://starsling.dev/ci-playbooks/anthropic#first-responder-agent)**: Use it when Failures on the default branch or in deploys wait for a person to open the logs, and each investigation starts from scratch. Check before adopting: A person stays on call and owns the fix. Start with failures on the default branch, keep the agent read-only, and compare its first analyses with what the person finds before widening it to deploys. Adapted for GitHub Actions; compare the original system with yours. ## Apply this technique ````text Evaluate and apply this CI technique to my repository: Put an agent on call to post an evidence-grounded analysis on every CI and CD failure. An agent joins the on-call channel, picks up each CI and CD failure, and posts an analysis grounded in the evidence it found for the person on call to review. Read the technique and its source-backed company examples: https://starsling.dev/ci-playbooks/techniques/first-responder-agent.md Inspect this repository’s CI workflows, build and test commands, dependencies, and recent run timings. Identify where this technique applies and explain the expected benefit using this repository’s evidence. Measure a baseline from recent comparable runs: elapsed time, queue time, runner minutes and failures where available. If data is unavailable, say what cannot be measured. Compare relevant company approaches, checking prerequisites, platform differences and tradeoffs. Rank applicable options by evidence, fit and measurable opportunity; do not present company results as predictions for this repo. Recommend the smallest safe change, show what checks and fallback paths remain, and ASK ME WHICH CHANGE TO APPROVE BEFORE EDITING ANY FILES. Only after approval, work on a new feature branch, make one reversible change, and open a reviewable pull request. Never merge or push to a default or protected branch. Keep required checks, test coverage, deployment and release behavior intact. Validate the change and compare equivalent runs where possible; report limitations and what should be monitored after merge. ```` ## How Anthropic used this technique Read what Anthropic changed and follow the original sources. ### Anthropic: Put an agent on call to post an evidence-grounded analysis on every CI and CD failure **Translated to GitHub Actions** - What Anthropic built: Anthropic made Claude Tag the on-call first responder, with an orchestration agent and executor subagents writing one situation report in the incident channel. - Your GitHub Actions version: A workflow that runs when a required check fails on the default branch or a deploy fails, reads the failed run's logs, the triggering commit and the recent runs of the same job, and posts its analysis for the person on call. Anthropic made Claude Tag the on-call first responder for CI and CD failures. When an alert is escalated to an incident, an orchestration agent spins up executor subagents that investigate each source of truth, then synthesizes what they find into one situation report in the incident channel. Standing instructions are skills committed in a GitHub repository, and a lessons file the agent appends to is read at the start of every investigation. In GitHub Actions, the same role is a workflow that runs when a required check fails on the default branch or a deploy fails, reads the failed run's logs, the triggering commit and the recent runs of the same job, and posts its analysis, with the lines and runs it relied on, for the person on call. Evidence: Company-reported result or observation Claude posts its first evidence-grounded analysis a median of 14 minutes after an incident opens, and in the fastest cases names the root cause within 4 minutes in its first report. ([Source](https://claude.com/resources/articles/ai-ci-cd-on-call)) Use it when: Failures on the default branch or in deploys wait for a person to open the logs, and each investigation starts from scratch. Before applying it: A person stays on call and owns the fix. Start with failures on the default branch, keep the agent read-only, and compare its first analyses with what the person finds before widening it to deploys. GitHub Actions adaptation: Compare the source system with your own workflow before applying this translation. [Read more in the Anthropic playbook](https://starsling.dev/ci-playbooks/anthropic#first-responder-agent) [Primary source](https://claude.com/resources/articles/ai-ci-cd-on-call) ## CI optimization is continuous work. Apply this technique yourself, or use StarSling Runners for faster GitHub Actions and optional agent-driven optimization PRs, which teams on eligible paid plans can turn on. [Get self-improving CI](https://starsling.dev/products/runners) ## Company Playbooks ### [Anthropic](https://starsling.dev/ci-playbooks/anthropic) Run only affected tests ### [DoorDash](https://starsling.dev/ci-playbooks/doordash) Run build and test in parallel, sharded across CI agents ### [Linear](https://starsling.dev/ci-playbooks/linear) Benchmark the same job on faster hardware ### [Mastra](https://starsling.dev/ci-playbooks/mastra) StarSling runners for the slowest test workflow ### [OpenAI](https://starsling.dev/ci-playbooks/openai) Rank CI fixes by the minutes they save ### [Ramp](https://starsling.dev/ci-playbooks/ramp) Move a CPU-bound test suite to a faster CPU ### [Shopify](https://starsling.dev/ci-playbooks/shopify) Let an agent iterate against one CI metric ### [Spotify](https://starsling.dev/ci-playbooks/spotify) Build arm64 on native arm64 runners ### [Stripe](https://starsling.dev/ci-playbooks/stripe) Run only the tests a change can reach ### [Uber](https://starsling.dev/ci-playbooks/uber) Build the merge-queue speculations most likely to be needed