Bernstein is an open-source governance layer for AI agents that applies written policies, schedules agent work deterministically, controls approvals, and records verifiable execution history. It is used to coordinate CLI coding agents and other agent workloads, including parallel runs and offline review, while the catalogue entries provide its agent workflows and controls.
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 skills add sipyourdrink-ltd/bernstein --skill bernstein-rungit clone --depth 1 https://github.com/sipyourdrink-ltd/bernsteinWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/sipyourdrink-ltd/bernstein/bernstein-run)<a href="https://agentmods.dev/skills/sipyourdrink-ltd/bernstein/bernstein-run"><img src="https://agentmods.dev/badge/skills/sipyourdrink-ltd/bernstein/bernstein-run/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/sipyourdrink-ltd/bernstein/bernstein-run"><img src="https://agentmods.dev/badge/skills/sipyourdrink-ltd/bernstein/bernstein-run.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What 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 5.1 | $0.00075 | $0.00675 |
| Opus 5.5 | $0.00030 | $0.00270 |
| Sonnet 5.5 | $0.00015 | $0.00135 |
| Haiku 4.5 | $0.00007 | $0.00068 |
Grade A, and why
bernstein-run 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 1mo ago.
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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run a verified multi-agent goal
Bernstein is the open-source governance layer for AI agents, with a deterministic multi-agent scheduler. This skill wraps the three-step operator loop: submit a goal, monitor progress, and verify the finished run against its Merkle-chained lineage journal and HMAC audit chain.
Prerequisites
bernsteinon PATH (pip install bernsteinoruv tool install bernstein).- Run commands from the project root (the directory holding
.sdd/after the first run). - First-time setup:
bernstein initscaffolds config and templates.
Workflow
1. Submit
bernstein run -g "<goal>"
- The goal is a plain-English description of what to build or fix.
- Bernstein plans tasks, assigns roles (backend, frontend, qa, security, devops), and spawns agents in isolated git worktrees.
- Preview without spending:
bernstein run --dry-run -g "<goal>".
2. Monitor
bernstein status
bernstein cost
statusshows task summary and agent health.costshows the spend so far against configured budgets.
3. Verify
bernstein lineage verify <run_id>
bernstein audit verify
lineage verifyrecomputes the run's Merkle hash chain and every HMAC tag; exit 0 means the chain is intact and non-empty.audit verifychecks the HMAC audit chain and Merkle tree across the whole workspace; run it before trusting or shipping the result.
Report the run as complete only when both verifications exit 0.
Verifying this skill's own install
Installs of this skill are receipt-backed. To confirm bernstein is on PATH
and the bundled skill asset resolves in this session:
bernstein skills package show
To prove the skill content the session is driving matches what was installed:
bernstein skills package verify --dest <path-to-this-skill-directory>
Exit 0 means the installed tree's content address matches a receipt
anchored in the lineage spine and audit chain. These two commands are the
self-check contract the multi-host conformance sweep replays per agent host
(bernstein skills package conformance).
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago Changed 7ae7501bd50c
- 1mo ago First seen · 91 lines · 75 tokens per session scan A eb3798b7b841
bernstein-run is a skill published in the GitHub repository sipyourdrink-ltd/bernstein (1,260 stars, last pushed today), licensed Apache-2.0. It adds 75 tokens to every session and 675 once invoked, about $0.0003 per session on Opus 5.5. 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
Agent Audit Trail
Append-only, hash-chained audit log for AI agents. Records agent actions, tool calls, decisions, and external writes with provenance, timestamps, and sha256 chain integrity. Designed for compliance with EU AI Act Article 12 automatic event recording requirements for high-risk AI systems.
akf-trust-metadata
The AI native file format. EXIF for AI — stamps every file with trust scores, source provenance, and compliance metadata. Embeds into 20+ formats (DOCX, PDF, images, code). EU AI Act, SOX, HIPAA auditing.
data-cleaning
Use when a raw table is too dirty to trust — nulls, sentinels, duplicate rows, category sprawl, mixed types, bad dates — and you need a re-runnable clean() plus a schema gate that fails loud. NOT emitting .xlsx (that is spreadsheet-ops), NOT acquiring rows (that is data-scraper), NOT parsing PDF/HTML into rows (that…
ai-research-reproduction
Rigor Reproduce compatible skill slug for README-first deep learning repository reproduction. Use when the user wants an end-to-end, minimal-trustworthy flow that reads the repository first, selects the smallest documented inference or evaluation target, coordinates intake, setup, trusted execution, optional trusted…
martinloop-govern
Govern coding-agent implementation with MartinLoop budgets, preflight checks, stop conditions, recovery, and verifier-backed evidence.
techpack-creator
Scan a repository that already has Claude Code configuration (hooks, skills, CLAUDE.md, settings, MCP servers) and package it into a valid MCS techpack.yaml for distribution via mcs. Use this skill whenever the user wants to create a tech pack from an existing repo, convert a Claude Code setup into a shareable MCS…