authentik
authentik 2026.8 ships: Actors, domain-joined Agents, and a push past browser-mediated SSO
A side-by-side editorial comparison of Prowler and webpack — release velocity, themes, recent moves, and the top alternatives to consider.
Prowler's assistant decides what to do with findings; the patches keep the scanner honest
Prowler is shipping a minor release roughly weekly with patches filling the gaps. The agentic layer, Lighthouse, has moved from explaining findings to acting on them, with named skills attached to individual findings and every write path bound to the asking user's RBAC. Alongside that, 5.39.1 fixes an install path that had been quietly broken: 5.38.0 declared a cryptography floor its own dependencies capped below, so pip install prowler silently resolved back to 5.37.1.
webpack is absorbing its own loader ecosystem — TypeScript, CSS and HTML now build natively.
webpack 5.109 flips experiments.typescript, experiments.css, experiments.html and experiments.asyncWebAssembly to "auto", so a project on Node 22.6+ compiles TypeScript, CSS and HTML with no loader registered at all. The same release adds built-in build progress via infrastructureLogging.progress, resource hints (preload/prefetch/modulepreload/preconnect) on by default for ESM output, and CommonJS module concatenation. Underneath, the 5.108 line is a sustained parser-performance campaign — struct-of-arrays HTML ASTs, dropped unused CSS value tokens, lower peak memory.
Prowler is shipping a minor release roughly weekly with patches filling the gaps. The agentic layer, Lighthouse, has moved from explaining findings to acting on them, with named skills attached to individual findings and every write path bound to the asking user's RBAC. Alongside that, 5.39.1 fixes an install path that had been quietly broken: 5.38.0 declared a cryptography floor its own dependencies capped below, so pip install prowler silently resolved back to 5.37.1.
Two tracks run in parallel and rarely overlap. The minor releases push the commercial agentic surface forward — triage skills, page context, the MCP tool set — while the patches defend the parts everyone uses: dependency resolution, container CVEs, and check correctness. That second track matters more than its version numbers suggest, because a security scanner reporting PASS when an API call failed is worse than one that errors. 5.39.1 fixes exactly that in the ECS task-definition checks, and makes the SES public-access check evaluate every identity policy rather than stopping at the first.
Expect the next minor to extend Lighthouse skills to groups of findings rather than one at a time, with patch releases continuing to absorb Trivy and base-image CVE churn.
webpack 5.109 flips experiments.typescript, experiments.css, experiments.html and experiments.asyncWebAssembly to "auto", so a project on Node 22.6+ compiles TypeScript, CSS and HTML with no loader registered at all. The same release adds built-in build progress via infrastructureLogging.progress, resource hints (preload/prefetch/modulepreload/preconnect) on by default for ESM output, and CommonJS module concatenation. Underneath, the 5.108 line is a sustained parser-performance campaign — struct-of-arrays HTML ASTs, dropped unused CSS value tokens, lower peak memory.
The direction is unmistakable: webpack is deleting the reason its configs became notorious. Every capability that used to require ts-loader, css-loader or html-webpack-plugin is being pulled into core and defaulted on, which is the only credible answer to zero-config rivals that never asked users to assemble a loader chain. Alongside that, a bun target and repeated Bun/Deno compatibility guards show webpack no longer assumes Node is the only runtime it must run under.
Expect the "auto" experiments to graduate out of the experiments namespace into stable top-level options, and expect the loader-shaped surface area to keep shrinking as more of the pipeline lands in core.
Other Infra & APIs products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either Prowler or webpack.
authentik 2026.8 ships: Actors, domain-joined Agents, and a push past browser-mediated SSO
Rancher's public feed is a build-tag stream: three branches bumped the same Go image on one afternoon
Buildkite keeps converting hand-rolled agent workarounds into first-class CI primitives.
Cursor's agents stop waiting to be asked - they subscribe, and they hold a goal until it's done.
Nexus does the diagnosis; the agent is now reaching into the status page too.
Warp turned its quarter of software-factory essays into infrastructure you can buy.
See all Prowler alternatives → · See all webpack alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
They serve adjacent needs but don't currently overlap on shipped themes. Prowler is currently shipping more aggressively (velocity 7.5 vs 6.3), with 2 editorial sparks in the last 30 days against 1. See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. Prowler is currently shipping more aggressively (velocity 7.5 vs 6.3), with 2 editorial sparks in the last 30 days against 1. For your specific use case, the alternatives sections above list other Infra & APIs products to evaluate alongside.
Top Prowler alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "Prowler alternatives" section above for the current picks, or visit /alternatives/prowler for the full list with editorial commentary on each.
Top webpack alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "webpack alternatives" section above for the current picks, or visit /alternatives/webpack for the full list with editorial commentary on each.