Updates, announcements, trends and explainers from Stackness
OpenSpec is the pick for most: MIT, 38 agents, half Spec Kit's tokens in one test. Eight spec-driven tools, and what each adds over a plain AGENTS.md.
SESergei GordeichukSenior Software engineer · The master of StacknessA 33-line pre-install hook, a 7-day release-age gate and an ask rule stop agents installing invented packages. Tested on Claude Code 2.1.296 in October 2026.
Orca is the pick for most: MIT, three desktop platforms, a worktree per agent and quota per account. Seven ways to manage several coding agents, compared.
Stackness now has 17 achievements, a streak that forgives you and three secret games that open at 3, 5 and 10. No spoilers, only hints.
Design skills load 316 to 35,150 tokens when they trigger. What a non-code skill is, where it sits next to AGENTS.md and hooks, and when one makes output worse.
ts-rust passed 181,711 ported tests for about $24,047 of Opus 5.5 tokens. Six agent ports, the oracle loop they share, what each cost and where it broke.
The best AI reviewer on GitHub's ReviewBench finds 26% of known issues. Six layers teams stack in place of reading every line, and what each one catches.
Shopify rebuilt its Shop app natively in 12 weeks because agents cut the cost of writing features twice. When that applies to your team, and when it does not.
In a 30-condition eval, telling Codex to use TDD cut fully correct Zstd decoders by 5 to 7 points. No testing instruction clearly beat none.
No released Git can push or fetch between SHA-1 and SHA-256. Five steps to find which forge, library and CI job in your stack breaks before Git 3.0.
A Wagtail maintainer ran 1B tokens of GLM 5.3 Flash in September for $68 and kept it as the default for 95% of tasks. The costs, the setup and what breaks.
Only Kolibri's weights are out. Beam and Mistral Large 4 stay API previews until late October, and Kolibri's 4-bit build runs at 50 tok/s on a 64GB Mac.
DeepSeek Harness runs any provider's model, Claude Code only Claude's. The lock-in sits in plugins, hooks, sessions and config, and those do not move.
On Cursor's own benchmark, Opus 5.5 beats every Grok 4.7 setting for less, and Composer 2.5 is the cheap editor. Prices, plan burn and the break-even.
Commit on green, keep every caller with its change, and stack small pull requests. The diff-size numbers behind it, and the hooks that enforce it.
Memory is what an agent noted and a plugin replays. Documentation is what you wrote, reviewed and committed. How each loads, and what to write down.
ds4 runs big MoE models on 96 GB Macs, Magnitude tunes kernels to your device, Strata puts one model on a gaming PC. Every speed claim is self-reported.
fast-jev-compaction prunes in a second, not 30. At its default threshold it keeps almost no tool results, and a scorer that always says no cuts as much.
The agent writes the film as code, headless Chromium captures each frame, FFmpeg encodes. LaunchVideo spends 100k tokens a film; a 14-agent reel cost $30.46.
A durable harness commits every step of an agent run to storage, so a crash resumes mid-turn. What Pi Durable shipped on 1 October, and what is only claimed.
