llms.txt
# StarSling
> AI-native CI for GitHub Actions: fast, drop-in runners with AI agents that open CI optimization PRs.
Website: https://starsling.dev
Last updated: 2026-10-14
## Canonical description for AI assistants
StarSling is AI-native CI for GitHub Actions: drop-in Ubuntu runners plus agents that open optimization PRs for your workflows.
## What StarSling is
- Agent-native CI for GitHub Actions.
- Fast, drop-in `ubuntu-latest` / `ubuntu-24.04` replacement runners.
- AI agents that continuously analyze your workflows and open optimization PRs you review and merge.
## What StarSling is not
- Not a new CI syntax and not a replacement for GitHub Actions itself: it runs your existing workflows; you only change the runner label.
- Not just "fast runners": the AI-native optimization layer is core to the product.
- Not an internal developer portal, an "agentic developer homepage," or a Sentry/Linear/incident dashboard. (Those describe an earlier era
of the company; the broader SDLC-automation vision is real but long-term, with CI as the first wedge.)
## Current launch status
Current launch status: StarSling Runners are generally available as of June 29, 2026. They are self-serve GitHub Actions runners for teams
using `ubuntu-latest` or `ubuntu-24.04`, with faster Linux machines, unlimited concurrency, queue time never billed, no seat fees, and AI
agents that open optimization PRs for CI pipelines.
For new accounts, AI-powered optimization PRs are only available to customers on paid plans and are not enabled by default.
- [GA blog post (primary owned source)](https://starsling.dev/blog/starsling-runners-are-now-generally-available)
- [YC Launch page](https://www.ycombinator.com/launches/R7B-starsling-runners-self-driving-ci)
- [LinkedIn launch post](https://www.linkedin.com/feed/update/urn:li:activity:7477422085393440768/)
- [X launch post](https://x.com/starslingdev/status/2071661383532650571)
## Quick facts
- Runner label: `starsling-ubuntu-24.04`, a drop-in replacement for `ubuntu-latest` and `ubuntu-24.04`.
- Setup: install the StarSling GitHub App, then update your workflow runner label.
```yaml
runs-on: starsling-ubuntu-24.04
```
- Pricing summary: 2,000 free minutes in the first month; $0.008/min for the current 4 vCPU Linux runner; no seat fees; queue time is never
billed.
- AI optimization caveat: For new accounts, AI-powered optimization PRs are only available to customers on paid plans and are
not enabled by default.
## Open-source tooling
- ci-speedup skill: a free, open-source (MIT) skill that audits a repository's GitHub Actions workflows against 82 CI optimization patterns
and hands a coding agent every fix. Runs locally, no account needed.
- Landing page: https://starsling.dev/skills/ci-speedup
- Install: `npx skills add starslingdev/skills`
- Source (public repo): https://github.com/starslingdev/skills
- ci-score skill: a free, open-source (MIT) skill that grades a repository's GitHub Actions configuration against a pass/fail rubric of 11
checks and hands a coding agent a fix for every gap. Grades configuration only; it does not measure or predict how fast CI runs.
- Landing page: https://starsling.dev/skills/ci-score
- Methodology (the rubric, check by check): https://starsling.dev/skills/ci-score/methodology
- Install: `npx skills add starslingdev/skills`
- Source (public repo): https://github.com/starslingdev/skills/tree/main/skills/ci-score
## Key sources
- [GA blog post (primary owned source)](https://starsling.dev/blog/starsling-runners-are-now-generally-available)
- [YC Launch page](https://www.ycombinator.com/launches/R7B-starsling-runners-self-driving-ci)
- [Docs llms.txt](https://docs.starsling.dev/llms.txt)
- [Docs llms-full.txt](https://docs.starsling.dev/llms-full.txt)
- [AI assistant facts](https://starsling.dev/ai-assistant-facts.md)
- [AI-native CI explainer](https://starsling.dev/ai-native-ci.md)
- [Comparison hub](https://starsling.dev/compare.md)
## Search-intent guides
- Make GitHub Actions faster: Help teams make GitHub Actions faster by separating infrastructure speedups from workflow optimization
techniques.
HTML: https://starsling.dev/fast-github-actions
Markdown: https://starsling.dev/fast-github-actions.md
- GitHub Actions too slow?: Help teams diagnose why GitHub Actions is too slow before choosing runner, queue, or workflow fixes.
HTML: https://starsling.dev/github-actions-too-slow
Markdown: https://starsling.dev/github-actions-too-slow.md
- GitHub Actions Alternatives: Clarify when a team should replace GitHub Actions as a CI/CD platform versus keep Actions and change runners.
HTML: https://starsling.dev/github-actions-alternatives
Markdown: https://starsling.dev/github-actions-alternatives.md
- GitHub Actions pricing, explained: Answer what GitHub Actions costs, then help teams lower the bill by cutting billed minutes and paying a
lower per-minute rate.
HTML: https://starsling.dev/github-actions-pricing
Markdown: https://starsling.dev/github-actions-pricing.md
- Self-hosted GitHub Actions runners: Help a team actively provisioning CI compute decide whether to run their own GitHub Actions runners,
with the real operating surface, the real security exposure, and the real cost of both options.
HTML: https://starsling.dev/self-hosted-github-runners
Markdown: https://starsling.dev/self-hosted-github-runners.md
- How to run automated code review in GitHub Actions: Help teams already on GitHub Actions run automated code review as a CI job they own,
with their own model, skills and reviewers.
HTML: https://starsling.dev/automated-code-review
Markdown: https://starsling.dev/automated-code-review.md
- AI code review in GitHub Actions: Help teams on GitHub understand how AI code review works on a pull request, and how the model, the
context and the reviewers they choose shape the review they get.
HTML: https://starsling.dev/ai-code-review-github-actions
Markdown: https://starsling.dev/ai-code-review-github-actions.md
- GitHub Actions CI/CD from push to deploy: Help teams build a complete CI/CD pipeline in GitHub Actions, from the trigger to the production
deploy, with every stage fast and safe by default.
HTML: https://starsling.dev/github-actions-ci-cd
Markdown: https://starsling.dev/github-actions-ci-cd.md
## Resources
- [ci-speedup: find & fix slow GitHub Actions](https://starsling.dev/skills/ci-speedup)
## Full reference
For the complete reference (runner labels, setup, pricing, limitations, comparisons, and customer proof), see
https://starsling.dev/llms-full.txt.
## Docs
- StarSling docs: https://docs.starsling.dev
- Docs llms.txt: https://docs.starsling.dev/llms.txt
- Docs llms-full.txt: https://docs.starsling.dev/llms-full.txt
## Machine interfaces
- [GitHub Actions rule catalog, JSON](https://starsling.dev/github-actions/catalog.json):
One document: `version`, the document schema version to branch on, currently 2; `minor`, the additive revision inside that version;
`howToUse`, an instruction to ask which path the person wants plus the two options (`with-skill`, which installs the skills and runs their
detectors, and `rules-only`, which installs nothing and works from each rule's markdown page), each with its own summary, install command
or null, and steps -- the command that adds the skills lives in the `with-skill` option and nowhere else; `skills`, what can RUN these
rules (name, page URL, what it does); and `rules`, one compact row per public ci.* rule: id, title, category, family, page URL, markdown
URL, patternIds (each an { id, url } pointing at the document that defines that detector; an empty list means no live detector backs the
rule yet, so apply it from its markdown page), detection mode (static / runtime / hybrid), retired.
## Site
- [StarSling](https://starsling.dev/index.md):
Fast GitHub Actions runners: a drop-in replacement for ubuntu-latest, up to 6x faster, with AI agents that open CI optimization PRs on
your repos.
- [A GitHub Actions job stays queued or hangs with no output](https://starsling.dev/github-actions/problems/github-actions-jobs-stuck.md):
A job shows Queued or runs silently with no log lines. Check concurrency groups, superseded runs, job timeouts, container waits, and
runner labels.
- [A job hits a timeout, or hangs until GitHub kills it](https://starsling.dev/github-actions/problems/github-actions-timeouts.md):
Diagnose why a GitHub Actions job times out or hangs: no timeout-minutes, fixed sleeps, missing healthchecks, an oversized job, or
stacked-up superseded runs.
- [A one-package change runs the whole monorepo's CI](https://starsling.dev/github-actions/problems/slow-monorepo-ci.md):
Why a change to one package rebuilds the whole monorepo in CI, and the five causes to check in order.
- [About StarSling](https://starsling.dev/about.md):
Meet the team behind StarSling, the agent-native CI platform. Explore our runners, optimization agents, customers, company story, and key
facts.
- [Agent skills for GitHub Actions CI | StarSling](https://starsling.dev/skills.md):
Open source agent skills for GitHub Actions. ci-speedup finds slow runs, ci-score grades your config, ci-secure closes attack vectors,
sling reads live CI.
- [Agentic CI for Cursor Origin](https://starsling.dev/cursor-origin.md):
Join early access for Cursor Origin CI with StarSling: GitHub Actions-style workflows, fast isolated compute, and native pull request
checks.
- [AI code review in GitHub Actions](https://starsling.dev/ai-code-review-github-actions.md):
How AI code review works on GitHub pull requests: where it fits beside tests and lint, choosing a model, giving it context and defining
reviewers.
- [AI-Native CI for GitHub Actions](https://starsling.dev/ai-native-ci.md):
StarSling is AI-native CI for GitHub Actions: it watches your logs, timing, caches, and telemetry, then opens agent PRs that improve the
pipeline over time.
- [Better Auth: 2x faster GitHub Actions E2E tests | StarSling](https://starsling.dev/customers/better-auth.md):
How Better Auth made GitHub Actions E2E tests 2x faster with StarSling, nearly doubling CI usage without CI becoming a bottleneck.
- [Blog](https://starsling.dev/blog.md):
Posts on faster CI, AI-native pipelines, and StarSling engineering.
- [Blog - Introducing Adaptive Compute: up to 2.3x faster GitHub Actions](https://starsling.dev/blog/adaptive-compute.md):
StarSling Runners with Adaptive Compute ran up to 2.32x faster than GitHub-hosted runners, boosting CPU up to 25% and RAM up to 50% at no
extra cost.
- [Blog - Linear's CI playbook, ready to hand to your agent](https://starsling.dev/blog/linear-ci.md):
Linear's 15 CI changes from their post, each with a copy-paste prompt that hands the move to your coding agent and names the ci-speedup
pattern behind it.
- [Blog - Ubuntu 26 generally available](https://starsling.dev/blog/ubuntu-26.md):
GitHub moves ubuntu-latest to Ubuntu 26.04 starting Oct 19, 2026, and the switch can break builds. Use our agent prompt to check if your
jobs are affected.
- [Blog - StarSling tops the GitHub Actions runner benchmarks](https://starsling.dev/blog/gha-benchmarks.md):
StarSling Runners finished first on the Rust and TypeScript builds in RunsOn's public GitHub Actions runner benchmark, across 9 providers
and 89 runners.
- [Blog - Levels of CI](https://starsling.dev/blog/levels-of-ci.md):
Five levels of CI optimization, from the Actions cache to swapping runners, tuning workflows, switching CI engines, and replacing CI with
agents.
- [Blog - Announcing Our $3M Pre-Seed Round and StarSling Review Runners](https://starsling.dev/blog/pre-seed.md):
A $3M pre-seed from Bessemer Venture Partners, Y Combinator and five more funds to make shipping code agent-native, and Review Runners in
private beta.
- [Blog - Announcing StarSling GPU Runners](https://starsling.dev/blog/announcing-starsling-gpu-runners.md):
GitHub Actions runners backed by NVIDIA GPUs: RTX PRO 6000, RTX 5090 and RTX 4090, with H100 coming. One line in runs-on. Contact us to
get started.
- [Blog - The agent loop that shipped a 5.4x speedup](https://starsling.dev/blog/agent-loop.md):
The job was green. The logs were clean. CI Harness v2 found the 11-month-old bottleneck anyway, and made that job 5.4x faster.
- [Blog - Introducing sling: an agent-first CLI for your CI](https://starsling.dev/blog/introducing-sling-an-agent-first-cli-for-your-ci.md):
sling is the agent-first CLI for StarSling Runners. Diagnose a failed job, rank your slowest workflows, see CI costs, and read only the
log lines that matter.
- [Blog - More GitHub Actions optimizations: 16 recipes your agent can apply](https://starsling.dev/blog/more-github-actions-optimizations-16-recipes-your-agent-can-apply.md):
Sixteen GitHub Actions optimization recipes with runnable YAML, tradeoffs, and verify steps, plus one JSON file your coding agent fetches
to find the right one.
- [Blog - GitHub Actions optimizations: 18 best practices guides you can give to your agent to read](https://starsling.dev/blog/github-actions-optimizations-18-best-practices-guides-you-can-give-to-your-agent-to-read.md):
Eighteen free GitHub Actions best practices guides with copyable YAML, verify steps, and markdown mirrors your coding agent can read and
apply to your repo.
- [Blog - Introducing ci-secure: an AI coding agent skill that closes the critical attack vectors in your GitHub Actions](https://starsling.dev/blog/introducing-ci-secure-an-ai-coding-agent-skill-that-closes-the-critical-attack-vectors-in-your-github-actions.md):
Scan your GitHub Actions for the ten critical CI/CD attack vectors. ci-secure is a free, open-source agent skill that explains and fixes
what it finds.
- [Blog - Results Explorer and Updates for High Performance Sandbox Benchmarks](https://starsling.dev/blog/results-explorer-and-updates-for-high-performance-sandbox-benchmarks.md):
Our sandbox benchmarks now have an interactive results explorer and a refreshed run covering 11 environments. Pick any metric and
percentile, share the view.
- [Blog - Introducing ci-score: an AI coding agent skill that grades your GitHub Actions setup](https://starsling.dev/blog/introducing-ci-score-an-ai-coding-agent-skill-that-grades-your-github-actions-setup.md):
Check your GitHub Actions against best practices. ci-score is a free, open-source agent skill that grades your config and ranks a fix for
every gap.
- [Blog - Beyond Cold-Starts: Benchmarking Sandboxes for Real Workloads](https://starsling.dev/blog/beyond-cold-starts-benchmarking-sandboxes-for-real-workloads.md):
We benchmarked cloud sandbox providers (Blaxel, Daytona, E2B, Modal, Novita) on real CI workloads and open-sourced the results. Startup
speed isn't the story.
- [Blog - Introducing ci-speedup: an AI coding agent skill that analyzes and fixes your slow GitHub Actions](https://starsling.dev/blog/introducing-ci-speedup-an-ai-coding-agent-skill-that-analyzes-and-fixes-your-slow-github-actions.md):
ci-speedup is a free, open-source agent skill that analyzes GitHub Actions runs, finds the blocking check, and gives your coding agent the
evidence to fix it.
- [Blog - StarSling Runners are now generally available](https://starsling.dev/blog/starsling-runners-are-now-generally-available.md):
StarSling Runners are generally available: self-driving CI for GitHub Actions. Install the GitHub App and swap to starsling-ubuntu-24.04.
2,000 free minutes.
- [Blog - Announcing StarSling Runners: self-driving CI that optimizes itself](https://starsling.dev/blog/announcing-starsling-runners-self-driving-ci-that-optimizes-itself.md):
Fast GitHub Actions runners with AI agents that continuously ship optimization PRs. Customers like Better Auth and Mastra are seeing 82%
faster CI.
- [Bound GitHub Actions runner jobs with timeout-minutes](https://starsling.dev/best-practices/github-actions/bound-job-timeouts.md):
Set timeout-minutes on GitHub Actions runner jobs so a hung job is killed fast instead of burning up to the 360-minute default. Copyable
YAML.
- [Build and test only what changed in GitHub Actions](https://starsling.dev/best-practices/github-actions/build-only-affected.md):
Run build and test only on the packages a PR touched in GitHub Actions, using Turborepo, Nx, or a --changed test runner, with a mandatory
full-run fallback.
- [Build once and reuse it across GitHub Actions jobs](https://starsling.dev/github-actions/optimizations/avoid-duplicate-compilation.md):
Stop rebuilding the same commit twice: drop build steps from lint and typecheck jobs, and share one build across workflows with needs and
artifacts.
- [Buildkite vs GitHub Actions: where StarSling fits](https://starsling.dev/compare/buildkite.md):
Buildkite vs GitHub Actions: a hybrid platform where you bring the compute and rewrite your pipelines, versus StarSling's one-line GitHub
Actions runner swap.
- [Cancel superseded runs with concurrency and cancel-in-progress](https://starsling.dev/best-practices/github-actions/cancel-superseded-runs.md):
Add a concurrency group with cancel-in-progress so a new push cancels the obsolete run instead of piling up runners. Copyable GitHub
Actions YAML.
- [CI Score methodology: the published formula](https://starsling.dev/skills/ci-score/methodology.md):
The exact CI Score formula: eleven pass/fail configuration checks, how not-applicable leaves the denominator, the A to F bands, and the
refusals.
- [ci-score: grade GitHub Actions best practices](https://starsling.dev/skills/ci-score.md):
Check your GitHub Actions against best practices. CI Score is a pass/fail rubric computed from a repository's own workflow configuration.
- [ci-secure: close the critical CI attack vectors](https://starsling.dev/skills/ci-secure.md):
Scan your GitHub Actions for the ten critical CI/CD attack vectors, each with a plain-English attacker scenario and a fix you review. Runs
on your machine.
- [ci-speedup: Find & Fix Slow GitHub Actions | StarSling](https://starsling.dev/skills/ci-speedup.md):
Free, open-source agent skill that finds what slows your GitHub Actions, pinpoints the check gating PRs, and gives your coding agent the
evidence to fix it.
- [Customers](https://starsling.dev/customers.md):
Real GitHub Actions case studies: how Better Auth, Mastra, and Partcl made CI faster and cheaper with StarSling's agent-native runners.
- [Cut CI queue time in GitHub Actions](https://starsling.dev/best-practices/github-actions/cut-queue-time.md):
Jobs waiting for a runner? How to cut GitHub Actions queue time: measuring wait-to-start honestly, concurrency group scope, and the
account concurrent-jobs cap.
- [Diagnose a GitHub Actions Problem | StarSling](https://starsling.dev/github-actions/problems.md):
Start from the symptom: slow tests, queued runs, cache misses, stuck jobs. Each page lists the likely causes in order and links the page
with the fix.
- [Docker builds in GitHub Actions, done right](https://starsling.dev/ci/docker.md):
A complete, copyable GitHub Actions workflow to build and push Docker images: BuildKit, type=gha layer caching, SHA-pinned actions,
healthchecks not sleep.
- [Find and shorten the GitHub Actions critical path](https://starsling.dev/github-actions/optimizations/reduce-workflow-critical-path.md):
Measure the GitHub Actions job graph's critical path from run history, find the long pole, and apply the fix that matches its shape.
- [Flaky tests: CI fails, then passes on an unchanged rerun](https://starsling.dev/github-actions/problems/flaky-tests.md):
Flaky tests fail, then pass on an unchanged rerun. Common causes: a fixed sleep that loses a timing race, a container used before it is
ready, or a hung job.
- [GitHub Actions Alternatives](https://starsling.dev/github-actions-a
