sem-ai

Deploy via Semaphore promotions. Manage deployment targets, promote pipelines, deploy-and-wait.

Manage Semaphore infrastructure — secrets, notifications, agent types, scheduled tasks, artifacts.

Diagnose sem-ai plugin issues when the sem-ai binary isn't installed yet, and guide the user through binary install. Use when the user reports sem-ai not working, MCP tools missing, slash command not found, `sem-ai connect` failing, or sees "command not found: sem-ai".

Triage and FIX flaky tests on a Semaphore project end-to-end — find the worst offenders, pull history, locate the test in the code, diagnose the root cause, write a fix or justified quarantine, and verify by re-running. Use whenever the user wants to fix/investigate flaky tests, de-flake CI, reduce intermittent failures, asks "why is CI randomly red", wants to quarantine a flake, or after `sem-ai flaky list` surfaces offenders. Goes BEYOND detection (test-intelligence) to root-cause + code change + verification.

Diagnose and fix CI pipeline failures. Step-by-step debugging with sem-ai.

Translate a repo's GitHub Actions workflows into an equivalent Semaphore pipeline. ONLY covers the GHA→Semaphore mapping and conversion procedure; for Semaphore-side depth (cache CLI, test-results, blocks structure, sharding, promotions) defer to the linked skills. Use when the user asks to convert/port/migrate GitHub Actions to Semaphore, says "translate this workflow" or "convert ci.yml", or the repo has `.github/workflows/` and the user wants Semaphore instead. Can be invoked directly as `/sem-ai:gha-to-semaphore` (or via the broader `/sem-ai:init` orchestrator).

Initialize Semaphore CI/CD for the current repository — bootstrap the Semaphore project, write a working `.semaphore/semaphore.yml` (translating from GitHub Actions if present, or from scratch), wire required secrets, validate, and watch the first workflow. Applies Semaphore-side defaults automatically — agent image, toolbox CLIs, `test-results` epilogue rule, sharding heuristics — by routing through the linked skills. Use when the user wants to set up CI on Semaphore, says "initialize", "bootstrap CI", "prepare CI/CD for this project", "create a workflow on Semaphore", "make a `.semaphore` config", or runs `/sem-ai:init`.

Monitor Semaphore project health — pass rates, recent failures, deployment status, trends.

Explain how Semaphore pipelines are structured — blocks, tasks, jobs, dependencies, parallelism — and link to sem-ai commands for inspecting them. Use when the user asks about Semaphore pipeline structure, blocks, tasks, jobs, parallelism, dependencies, fan-out / fan-in, why a job is waiting, how to split work into parallel blocks, or where to put agent / prologue / epilogue / secrets / env_vars.

Manage Semaphore CI/CD via sem-ai. Use when the user asks about CI status, pipeline failures, test results, deployments, secrets, notifications, scheduled tasks, deployment targets, project health, or anything related to their Semaphore pipelines and workflows — e.g. "CI status", "pipeline failed", "why did CI fail", "deploy to staging", "rerun the pipeline", "what's flaky", "check the build", "show me the logs", "promote to production", "validate yaml".

Run CI commands against local changes in a real Semaphore environment. Use when the user wants to test code before pushing, run tests in CI env, or iterate fast on fixes — e.g. "run tests", "test before pushing", "try this in CI", "does this pass", "test my changes", "run the build", "check if tests pass", "spin up CI environment".

After pushing commits to a Semaphore-backed repo, find the triggered CI run and watch it to completion, then report pass/fail. Use right after `git push`, or when the user asks "did my push pass", "watch CI", "is the build green", "what happened after I pushed", "watch the pipeline".

The Semaphore toolbox — CLIs preinstalled in every job for caching, artifacts, language version switching, services, retries, and SSH debugging. Authoritative reference for the cache / artifact / retry / sem-version / sem-service / checkout commands. Use when an agent needs to reason about Semaphore toolbox commands, questions arise about cache CLI semantics, artifact push/pull, sem-version, sem-service, retry, or checkout, pipeline-writing decisions involve any of these CLIs, or debugging "command not found" / "command line argument" errors involving toolbox commands.

Publish JUnit test reports from Semaphore jobs and surface them in the UI's Test Reports tab. Covers the `test-results` CLI, why it must live in `epilogue` (not the test command itself), and per-framework JUnit configuration for Go, pytest, RSpec, Jest, Vitest, ExUnit, Java. Use when a pipeline writes JUnit / test reports, the Test Reports tab is empty after a failure, failures publish silently, the user asks about flaky test surfacing or per-framework JUnit setup, or any time `test-results publish` / `test-results gen-pipeline-report` is involved.

Pull and analyze test results from a Semaphore pipeline — failure detail, flaky-test detection, JUnit parsing. Use when a job failed and you need to see WHICH test broke without changing the pipeline config or re-running; when a junit-only reporter hides failure detail in the log; when you're about to add a stdout reporter just to debug — pull the published JUnit artifact instead via `sem-ai test summary` / `sem-ai test report` / `sem-ai artifact get`. Supports Go, pytest, rspec, minitest, jest, ExUnit.

Spin up a short-lived Semaphore testbox to check what's actually installed on an agent — tools, versions, toolbox functions, runtimes — when the official image docs don't answer the question. Use when an agent needs to confirm whether a tool/runtime/version is preinstalled on a Semaphore agent ("does the runner have X", "what version of Y ships on ubuntu2204/ubuntu2404", "is jq/yq/gh/aws on PATH", "what toolbox functions exist"), or before adding a CI step that installs a tool — verify it isn't already preinstalled.

Explain Semaphore promotions — chained pipelines for deploys, gates, and environment fan-out — and link to sem-ai commands for triggering and inspecting them. Use when the user asks about Semaphore promotions, deploys, deployment targets, auto-promote, parameterized promotions, staging vs production gates, how to chain pipelines, or environment fan-out.