Spend cap, transactional undo, and proof receipts for AI coding agents.
A lightweight open-source (MIT) safety harness that runs under Claude Code, Cursor, OpenCode, Codex, Antigravity, Cline, and fx.sh.
$ kineti spend status
total $0 of $50; entries 0; tripped=false
$ kineti spend log --stage build --model sonnet --tokens-in 400000 --tokens-out 200000
logged $4.2 (stage build); run total $4.2
$ kineti spend log --stage build --model opus --tokens-in 100000 --tokens-out 60000
kineti: SPEND BREAKER TRIPPED: stage build total $10.2 reached ceiling $9.5 (exit code 3)
$ kineti spend check
kineti: TRIPPED: stage build total $10.2 reached ceiling $9.5 (exit code 3)
$ kineti spend reset
kineti: reset requires --i-am-human (breakers are human-only) (exit code 2)
$ kineti spend reset --i-am-human
breaker reset by human (exit code 0)
Personalized agents must understand the full human context without hallucinating, overriding user statements, or quietly mutating goals when third-party systems push back:
Global, Domain (Health, Work, Finance, Schedule, Taste), and Relationship (person-to-person) scopes. Scoped facts never leak into generic or un-scoped queries.DirectlyKnown (explicit user ground truth) > ObservedPattern (behavioral patterns) > Inferred (hypotheses). Machine inferences are strictly blocked from overwriting explicit user statements.BaselineRule (e.g. Vegetarian) $\to$ PermittedException (e.g. Eats eggs) $\to$ Preference (e.g. Prefers low dairy) $\to$ SafetyCeiling (e.g. Peanut allergy).core-native/)RwLock snapshot design with thread-safe read paths and deferred epoch reclamation of retired instances. Latency figures are withheld until reproducible benchmark scripts land in the repo.docs/PLAN.md for rationale and test vectors).KinetiConnectorProtocol trait with consequence level gating (Trivial, Operational, HighConsequence) and single-use SHA-256 payload authorization tokens.bin/, src/)kineti-evidence.ts.bin/kineti-evidence.ts).src/harness/benchmark.ts are design targets, not measured results: the ALE 76.4% pass rate, SWE-bench 4.2 min MTTR, and $0.31 per-outcome numbers are goal constants for the evaluation program. They have not been produced by empirical runs.Prerequisite: The CLI is distributed on npm and runs on Node.js (>= 18), but requires Bun (>= 1.1) installed on the system for governance execution (
curl -fsSL https://bun.sh/install | bash).
# Install globally
npm install -g kineti
# Or run directly with npx
npx kineti --help
Run Kineti directly as an MCP governance server inside your AI editor to enforce spend caps ($50 ceiling), transactional SAGA undo, and cryptographic test verification.
Run Kineti's auto-configurator in your project directory:
kineti init
This automatically detects your installed hosts (Claude, Cursor, OpenCode, Codex, Antigravity, fx.sh, Gemini, Cline) and links project rules and MCP configurations.
Claude Code (CLI):
claude mcp add kineti npx -y kineti mcp
Cursor & Windsurf:
Add to your project's .cursor/mcp.json or Cursor Settings $\to$ Features $\to$ MCP:
{
"mcpServers": {
"kineti": {
"command": "npx",
"args": ["-y", "kineti", "mcp"]
}
}
}
OpenCode:
opencode mcp add kineti npx -y kineti mcp
Or add to ~/.config/opencode/opencode.jsonc.
Codex (CLI):
Add to ${CODEX_HOME:-$HOME/.codex}/config.toml:
[mcp.servers.kineti]
command = "npx"
args = ["-y", "kineti", "mcp"]
Google Antigravity & Gemini: Run the automated host configurator:
kineti init --host antigravity
Fx.sh: Install skill rules directly to your fx environment:
kineti init --host fx
Claude Desktop:
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"kineti": {
"command": "npx",
"args": ["-y", "kineti", "mcp"]
}
}
}
Cline / Roo Code: Add a new stdio MCP server in settings:
kinetinpx["-y", "kineti", "mcp"]# Start the visual companion dashboard (Apple HIG)
kineti companion
# Open http://127.0.0.1:8788
# Start the MCP governance server directly
kineti mcp
# Check spend circuit breaker status ($50 default ceiling)
kineti spend check
# Run tests and save cryptographic proof receipt
kineti test -- echo hello
# Verify cryptographic test evidence freshness
kineti test check --label test
# Run stage-agnostic CI verification
kineti ci
All 8 native crates are published on crates.io:
# Add native core nervous system to your Rust project
cargo add kineti-core
# Or install the native CLI binary directly
cargo install kineti-cli
# Run all 251 native Rust unit and integration tests locally
cargo test --manifest-path core-native/Cargo.toml
# Run the 5 native verification demo flows
cargo run --package kineti-cli -- test-all
kineti/
├── bin/ # TypeScript Governance CLI tools (single router: kineti.js)
│ ├── kineti.ts # CLI router source (spend, saga, evidence, companion, mcp, ci)
│ ├── kineti-spend.ts # Hardware spend circuit breaker ($50 ceiling, exit code 3)
│ ├── kineti-saga.ts # LIFO undo stack & transactional rollback
│ ├── kineti-evidence.ts # Delimited SHA-256 test proofs bound to git tree hashes
│ ├── kineti-companion.ts # Apple HIG visual companion server (loopback only)
│ └── kineti-mcp.ts # Model Context Protocol (MCP) server
├── core-native/ # Pure Rust Nervous System Workspace (8 crates, 0 external dependencies)
│ ├── Cargo.toml # Workspace manifest
│ └── crates/
│ ├── kineti-core/ # Double-buffered snapshots, HLC, Kernel, Commit Gate
│ ├── kineti-memory/ # Epistemic Engine, Multi-scope context, Tombstones
│ ├── kineti-reflex/ # Fast sensory classification & emoji reflexes
│ ├── kineti-connectors/ # Protocolized connectors & permission gating
│ ├── kineti-actions/ # Action execution & confirmation gates
│ ├── kineti-gateway/ # Messaging webhooks & bridge
│ ├── kineti-harness/ # Outcome Verification Tickets (OVT) & Shadow workspaces
│ └── kineti-cli/ # Standalone native CLI binary
├── src/ # Shared TypeScript libraries (governance, scheduler, security)
├── skills/ # 16 agent workflow skills (installed by setup.sh)
├── hosts/ # 10 editor configurations (Cursor, Claude, Antigravity, etc.)
├── hooks/ # 5 hook text blocks for host setup
├── public/ # Static documentation & legal assets
├── website/ # Kineti marketing & research portal (React + Vite)
├── docs/ # Architecture specifications & security reports
│ ├── PLAN.md # Canonical production architecture spec
│ ├── AUDIT.md # Audit summary & benchmarks
│ ├── SECURITY_REPORT.md # Security audit & origin gating analysis
│ └── TUTORIAL-first-run.md # Developer quickstart
├── tests/ # 168 TypeScript governance & causal test suites
├── .github/ # GitHub Actions CI, issue forms, PR template, Dependabot
├── AGENTS.md # Universal rules for AI agents in this repository
├── SECURITY.md # Vulnerability disclosure policy & SLAs
├── CONTRIBUTING.md # Development setup, testing, and DCO sign-off
├── LICENSE # MIT License
└── package.json # kineti@0.3.5 npm package manifest
Today Kineti provides local ledger tooling, spend circuit breakers, and evidence verification that agents invoke during their lifecycle (conventions in hooks/ plus self-reported proofs via kineti-spend.ts and kineti-evidence.ts). An in-line MCP proxy gate with deterministic allow/ask/deny interception is in active sprint.
#![forbid(unsafe_code)] enforced across all critical crates.$50.00 default cap per project, settable at mirror time ($1-$1000); tooling exits with code 3 at 95%.For full architecture deep-dives and research treatises, refer to docs/README.md.
If you are an AI agent working in this repo, read AGENTS.md first. It holds all rules, workflows, memory spec, and program references in one file.
Distributed under the MIT License. See LICENSE for more information.
Humans and agents follow the same steps in CONTRIBUTING.md: setup, tests with proof, money rules, and the ship checklist.
FAQs
Kineti: spend cap, undo, and test proof for AI coding agents
We found that kineti demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Company News
Socket is joining the OpenJS Security Stewardship Program to fund Node.js vulnerability research, maintainer remediation, and security releases.
Security News
Two compromised GitHub Actions were re-enabled with malicious tags intact, exposing thousands of downstream repositories to Mini Shai-Hulud.
Research
/Security News
A malicious Firefox extension fetches its payload after installation to evade detection, steal Google session cookies, and automate account takeover.