Brevet: Change Control for What AI Agents Learn

A governed evolution loop that makes agent learning promotable, auditable, and revocable.

AI agents now change their own behaviour while they work. They save memories, write themselves new skills and edit their own instructions. Many of these changes help. Yet none of them passes through the steps an organisation expects when a person changes how work is done. Nobody writes the change down, nobody approves it, and when it turns out to be wrong there is no earlier version to go back to.

Brevet adds those steps. It is a local-first Python runtime that wraps the agent you already have and runs its learning as a governed evolution loop. When an expert corrects the agent's draft, Brevet records the correction and its reason as an override. Offline, in the dream cycle, overrides that keep recurring become candidate capabilities: proposed rules and other learned behaviour, with no authority. At the dawn gate, a named human or mission group (the accountable review board) decides which candidates to promote. The overrides then replay as evals, and the conservative gate stops a release that makes either half of them worse. Promoted capabilities ship in a signed release, listed in capabilities.lock, and a capability that proves wrong can be recalled, with every release that shipped it flagged. Every step is recorded on a hash-linked evidence chain.

Agents propose deltas; evidence tests them; humans promote them; the runtime only ever executes signed versions.

Three questions Brevet answers

A concrete example

A quality reviewer at a pharmaceutical plant checks an agent's severity rating for each equipment problem. The agent rates pump vibration during cleaning as minor. She overrides it to major every time, because that vibration is an early sign of seal wear. After four overrides, the dream cycle proposes a candidate rule. At dawn her mission group promotes it, and release 0.2.0 ships with the rule in its capabilities.lock. Months later, engineers trace the vibration to a faulty sensor, so the mission group recalls the rule and Brevet flags release 0.2.0.

import brevet
from brevet.runner import EvalRunner

agent = brevet.wrap(triage_agent)          # wrap the agent you already have

# Work and override: the agent drafts; the reviewer corrects the draft and says why.
result = agent.run("Pump P-301: vibration high during cleaning",
                   task_family="equipment_triage")
agent.record_final(result.task_id, "severity: major",
                   participant="human:qa.reviewer@example.com",
                   rationale="Vibration during cleaning is an early sign of seal wear.",
                   tags=["vibration-during-cleaning"])

# Dream: once the same override keeps recurring, it becomes a candidate.
agent.dream()
candidates = agent.dawn()                   # the dawn queue
rule = next(c for c in candidates if c.kind == "prompt_rule")

# Dawn: a named mission group promotes it. A dream:* approver is rejected.
agent.dawn(decide=(rule.capability_id, "promote"),
           approver="mission_group:quality_team")

# Evals: replay the overrides as tests, before and after the change.
before = agent.evaluate()
#    ...update your agent so that it follows the promoted rule...
after = agent.evaluate()

# Release: refused unless the conservative gate passes.
check = EvalRunner.compare(before, after)
agent.release(to_version="0.2.0", channel="trial",
              approver="mission_group:quality_team",
              delta_in=check["delta_held_in"], delta_out=check["delta_held_out"])

# Recall: the rule proves wrong. Then verify the whole evidence chain.
agent.recall(rule.capability_id, reason="The vibration came from a faulty sensor.",
             issued_by="mission_group:quality_team")
agent.verify()

Quickstart

pip install brevet
brevet demo                  # the whole loop as one command
brevet playground            # step through the loop in your browser

Everything runs on your own machine, with no model or network connection. To run the example above, clone the repository and run python examples/pump_vibration.py. For a guided, clickable tour, open docs/demo.html in a browser.

Works with the agent you already have

brevet.wrap() recognises agents built with LangGraph, the Claude Agent SDK, DeepAgents, AutoGen, LlamaIndex, Pydantic AI, the Google Agent Development Kit, CrewAI and the OpenAI Agents SDK, and it accepts any Python function. Brevet never changes the agent it wraps. uvx brevet mcp offers the whole loop to any MCP client (ABOUT.md shows the setup), and examples/claude-cowork uses it to govern what Claude itself learns.

Project status

Brevet implements the whole loop and keeps every record, and a workspace can require every decision to be signed by its registered approvers. Some protections depend on the system you deploy it in, such as verifying who holds each key and anchoring the evidence chain outside the machine. ABOUT.md lists them, and the paper sets them out in full.

Learn more

• ABOUT.md: the seven stages, the authority ladder, supported frameworks, the MCP server and commands, how Brevet fits with CHAP and Metis, and how the repository is organised.
• GLOSSARY.md: every term, with its plain meaning first.
• SPEC.md: the rules any implementation must follow.
• BENCHMARK.md: the proposed governed-adaptation benchmark.

Citation

If you use Brevet in research, please cite the paper Brevet: Change Control for What Self-Evolving AI Agents Learn (Shahid, Suttie and Black, 2026). CITATION.cff gives the software citation.

License

Apache-2.0. See LICENSE. Brevet is a Brightbeam project.