open-standards/jouleclaw
Repository files (latest commit first)
Filename Latest commit message Latest commit date
Transaction Science 289d584b02
All checks were successful
standards workspaces build + test / which workspaces this change touches (pull_request) Successful in 15s
repository checks / lockfiles, tests in CI, workflow hygiene, spec and section references, public text (pull_request) Successful in 1m24s
joule-code conformance gate / Build + round-trip C1-C4 vectors (pull_request) Successful in 2m3s
jouleclaw public surface must stay documented / Public-surface audit (jouleclaw-rs) (pull_request) Successful in 58s
standards workspaces build + test / joule-code-rs (joule-code/joule-code-rs, joule-code-rs, linux-arm64, true) (pull_request) Successful in 10m11s
standards workspaces build + test / jcp-rs (jcp-conformance, jcp/jcp-rs, -p jcp-energy --features nvml, jcp-rs, linux) (pull_request) Successful in 1m52s
standards workspaces build + test / joule-ui-rs (joule-ui/joule-ui-rs, joule-ui-rs, linux) (pull_request) Successful in 1m11s
standards workspaces build + test / joulecontract-rs (joulecontract/joulecontract-rs, joulecontract-rs, linux-arm64, true) (pull_request) Successful in 4m31s
standards workspaces build + test / joulehook-rs (joulehook/joulehook-rs, joulehook-rs, linux) (pull_request) Successful in 1m14s
standards workspaces build + test / jouleclaw-rs (-nodebug, jouleclaw/jouleclaw-rs, -p jouleclaw-energy --features rapl,nvml, jouleclaw-rs, linux-arm64, -C target-feature=+fp16) (pull_request) Successful in 17m16s
standards workspaces build + test / mesh-rs (mesh/mesh-rs, -p mesh-engine-gguf -p mesh-openai --features mesh-engine-gguf/gguf,mesh-openai/server, mesh-rs, linux-arm64) (pull_request) Successful in 1m45s
standards workspaces build + test / proof-rs (proof-conformance, proof/proof-rs, proof-rs, linux, true) (pull_request) Successful in 1m37s
standards workspaces build + test / sandbox-rs (sandbox-conformance, sandbox/sandbox-rs, sandbox-rs, linux) (pull_request) Successful in 2m28s
standards workspaces build + test / macOS energy backends (pull_request) Successful in 7m41s
standards workspaces build + test / jouledesk-rs (jouledesk/jouledesk-rs, jouledesk-rs, macos) (pull_request) Successful in 1m54s
jcp: a ModelIn caveat says which model a grant authorizes, and the frame names the model it runs
A grant's capability names one resource by content address. When that
resource is an inference endpoint, the model it runs was chosen inside the
payload, and the decision point sees only the payload's hash. So a grant
could authorize "this endpoint" but never "this endpoint, running these
models". Only PayloadIs came close, and it makes the grant single-use. JCP
enforced a model's integrity (refuse wrong bytes) but not authorization
over which model runs.

- `Frame.model: Option<ContentId>` is the content address of the model the
  action runs. The dispatching runtime fills it, since it knows what it will
  load. It is omitted from the encoding when absent, so frames and every
  receipt and vector over them keep their bytes.
- `Caveat::ModelIn(Vec<ContentId>)` requires the frame to name one of these
  models. It fails closed: a frame naming no model is denied
  `caveat-unsatisfied`, so a dispatcher that does not say what it will run
  cannot exercise the grant. A delegate narrows the set by adding a smaller
  ModelIn beside the inherited one; both must hold, so the intersection
  governs. This is authorization by content address, not a registry
  (SPEC §12).
- SPEC §4.9 and §6 describe both, and C24's conformance check now covers
  ModelIn.
- Every Frame literal in the repository gains `model: None`. That is 28
  sites across jcp-rs, joule-code-rs, joule-ui-rs, jouleclaw-rs and
  sandbox-rs.

Checked on the build host:
- jcp-rs: 496 tests pass, and the conformance suite passes all checks,
  including C24 with ModelIn.
- Every vector gen-vectors writes is byte-identical to the published one.
- All eight workspaces that depend on jcp-schema pass `cargo check
  --all-targets`. The consumers jouleclaw-jcp, joule-code-jcp, sandbox-jcp
  and joule-ui pass their tests.
- Mutations: letting a frame with no model pass, and encoding an absent
  model, each fail tests.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-10-17 04:21:21 -04:00
..
conformance/v1 jouleclaw SPEC.md: normative §§8-12 for waves 12-21 2026-06-02 10:42:25 -04:00
jouleclaw-rs jcp: a ModelIn caveat says which model a grant authorizes, and the frame names the model it runs 2026-10-17 04:21:21 -04:00
scales compliance: publish the EU draft PUE bands as a sourced data pack, not as code 2026-09-06 06:24:25 -04:00
vocab jouleclaw: wave-24 — contract + handoff + mission + value report 2026-06-03 02:35:34 -04:00
LICENSE jouleclaw: standard skeleton + Phase 1 foundation crates 2026-05-26 09:55:11 -04:00
LICENSE-SPEC licences: ship the Apache-2.0 text the code already declares 2026-10-09 17:22:32 -04:00
README.md licences: ship the Apache-2.0 text the code already declares 2026-10-09 17:22:32 -04:00
ROADMAP.md docs: strip vanity metrics from README + ROADMAP 2026-06-02 07:30:50 -04:00
SPEC.md jouleclaw-energy: a process reads itself only once the reading thread has left its core 2026-10-12 20:50:52 -04:00

JouleClaw — Energy-Optimized AI Runtime

The substrate for AI that costs Joules, not tokens.

Capability per Joule, not capability per parameter.

What JouleClaw is

JouleClaw is a pure-Rust, omni-modal AI runtime + harness + agentic layer governed by Transaction Science as an open standard (5th in the open-standards quartet, alongside OpenPay, Smart Byte, EOC, and WAI).

Every operation JouleClaw performs walks an explicit energy-tiered cascade, top to bottom, stopping at the first stage that closes the query:

Tier Name Cost class What runs here
L0 Cache picojoules Content-addressed cache hits
L1 Lawful nanojoules Deterministic, pre-compiled primitive (text/code only)
L2 Embed sub-millijoules Matryoshka embeddings + nearest-neighbor + hybrid search
L3 Model joules Local SSM / ternary / multimodal / diffusion / etc.
L4 Wire tens of joules Remote frontier RPC (escape hatch only)

L1 (Lawful) is only meaningful for text and code modalities. For image, audio, video and 3D generation, the cascade collapses to L0 / L2 / L3 / L4 — the energy savings come from caching plus dispatch to the cheapest model that meets the quality bar.

Why JouleClaw exists

The mainstream "agentic AI" ecosystem treats inference as the entry point. Every state shift, every tool call, every routing decision runs through a 70-billion-parameter stochastic engine, burning kilojoules to do work that deterministic compute or a hot cache could resolve for microjoules.

JouleClaw inverts the default. Inference is the last resort, not the first. Frozen model weights are less trustworthy than current world state; fresh, provenance-stamped retrieval beats synthesis from training cuts. Every operation is metered in microjoules where the hardware permits it, and in millijoules where it doesn't — the spec is honest about the floor on every platform.

The thermodynamic circuit breaker is the safety mechanism. Set an energy budget; when measured consumption exceeds the budget, the breaker trips and the operation halts. Hallucination loops cannot survive an honest joule ledger.

What's here

Path Contents
SPEC.md JouleClaw v1 spec — cascade tiers, energy contract, receipt
CHARTER.md Stewardship pattern (shared across open-standards)
jouleclaw-rs/ Apache-2.0 Rust reference implementation
spec/ Tier specifications, energy provenance, receipt format
conformance/v1/ Signed wire vectors for cross-implementation verification

Reference implementation

jouleclaw-rs/ is the Apache-2.0 reference implementation in Rust. It runs on Apple Silicon (M3/M4/M5 with UMA + Metal 4 Tensor APIs), AMD Strix Halo / Ryzen AI Max, Intel Lunar Lake, NVIDIA Jetson, and discrete NVIDIA / AMD GPUs.

It is omni-modal: text, code, vision, audio, video, image diffusion, 3D Gaussian splatting, and cross-modal fusion all dispatch through the same energy-tiered cascade.

The MCP tool surface is a JCP consumer: jouleclaw-jcp wraps any jouleclaw-mcp metered tool so every dispatch is decided by the JCP runtime against a signed, metered Grant before a byte reaches the tool — the tool's identity is the content-addressed resource, the call's energy estimate is debited from the grant's joule budget, and every decision (allow and deny alike) seals as a JCR-1 receipt. An injected instruction can name an ungranted tool; it cannot make the gate dispatch it.

Status

v0.1.0 reference implementation in active development. The spec is draft status. Cross-implementation conformance vectors will land with v1.0.0.

License