Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/block/berd/experimental-featuresnpx skills add block/berd --skill experimental-featuresgit clone --depth 1 https://github.com/block/berdWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00019 | $0.01096 |
| Opus 5 | $0.00010 | $0.00548 |
| Sonnet 5 | $0.00004 | $0.00219 |
| Haiku 4.5 | $0.00002 | $0.00110 |
Grade A, and why
experimental-features scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Experimental Features
Use experiments for opt-in, user-local in-progress Berd UI or workflow behavior. Do not use experiments for secrets, credentials, backend authority, packaged policy, or app state that should survive graduation as a normal preference.
When To Use Experiments
- An individual user opts into unstable UI or workflow behavior.
- Stable behavior can remain the default path.
- Config is small, non-sensitive, typed, and user-editable.
- The feature can be graduated or removed later.
When To Use distro.json
Use distro.json for packaged build policy and startup defaults, especially
when the Tauri shell or sidecar needs bundled resources/config.
Good distro fits include providerAllowlist, kgoose,
featureToggles.costTracking, bundled config.yaml, bin/, skills/, and
agents/.
Do not use distro.json for normal app state, user preferences, dynamic runtime
switches, ACP-backed data, or per-user experiments.
Registry Shape
Add experiments only in
src/features/experiments/experimentDefinitions.ts.
Each definition needs:
id: stable kebab-case stringtitleKeyanddescriptionKey: settings i18n keysconfig: optional typed controls
Experiments without a manual per-experiment override follow the global
autoEnable preference. That preference defaults on in dev builds and off in
production builds. Users can force an experiment on/off or reset it back to auto
from settings.
Config entries under an experiment are settings for that experiment, not nested experiments or independent feature flags. Keep them stored with the parent experiment and gate their runtime effect on the parent experiment being enabled.
Supported config controls:
boolean: switch with a boolean defaultselect: fixed string options with a defaultnumber: default plus optional min/max/steptext: default plus optional placeholder; never for secrets
Use getExperiment(id) or useExperiment(id) for callers. When an experiment is
disabled, keep config stored but gate behavior as disabled.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 140 lines · 19 tokens per session scan A 2b7bc4048840
experimental-features is a skill published in the GitHub repository block/berd (828 stars, last pushed 2d ago), licensed Apache-2.0. It adds 19 tokens to every session and 1,096 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…