| Filename | Latest commit message | Latest commit date |
|---|---|---|
The repository tracks the latest stable Rust, pinned at the root. A
`rust-version` key states a *floor* -- the oldest toolchain a crate supports --
which is a different promise, and one this repository does not make. It also
said 1.97 while the pin was 1.98.0 and nothing verified it, so it was a claim in
the wrong direction and unchecked in both.
So it goes, everywhere:
59 rust-version = "1.97" (workspace and package declarations)
417 rust-version.workspace = true (members inheriting them)
476 files, 476 deletions, 0 insertions
The 417 are why this is not the sixty-file change it looks like from a grep for
the value. A member inheriting a key that no longer exists does not fall back to
a default -- its manifest stops resolving. Removing the declarations without the
inherits would have broken every workspace that has members.
Deliberately untouched: `joule-code-rs/vendor/wtransport/Cargo.toml`. That is a
vendored third-party crate declaring a floor about itself, not about us, and our
pin is above it. Editing vendored source to satisfy our own policy would make
the copy diverge from upstream for no benefit.
Verified rather than assumed:
- all 493 `Cargo.toml` files in the tree parse (`tomllib`)
- `cargo metadata --no-deps --locked` resolves on `joule-code-rs`, `wai-rs`,
`openpay`, `mesh-rs`, `sandbox-rs` and `jcp-rs` -- the check that actually
exercises workspace inheritance, which a TOML parse does not
- no `Cargo.lock` changed; `rust-version` is not lockfile state
`wai/wai-verify/Cargo.toml` already carried a comment saying it omits
`rust-version` on purpose, so this makes the rest of the tree agree with the one
crate that had already decided.
What this gives up, stated plainly: cargo can no longer greet an older
toolchain with "requires rustc 1.97" and will instead fail wherever the code
first needs something newer. That is the cost of not maintaining a floor, and
it is the right cost when the floor was fiction.
Co-Authored-By: Claude Opus 5 <[email protected]>
|
||
| .. | ||
| conformance/v1 | ||
| joulehook-rs | ||
| LICENSE | ||
| LICENSE-SPEC | ||
| README.md | ||
| SPEC.md | ||
JouleHook — Metered, Gated, Attested Agent Work in Any Host
Provisional name — the steward may rename; the profile string joulehook/1 and the
seam names are the stable surface.
The energy-accounting and methodology-gating you build for one coding agent is worth little if it's trapped in that one agent. JouleHook carries it into the editors developers already use — Claude Code, Cursor, Windsurf, Antigravity, Zed, JetBrains, OpenCode — without forking any of them.
It rests on one finding: no host lets an extension observe its agent's tokens or intercept its native tools — but every host exposes the same three seams, and JouleHook is the normative contract over them:
A per-action gate, a model-call meter, and a receipt surface. Bind those three to a decide-outside-the-model grant check, an honest energy meter, and a sealed receipt, and any agent's work becomes a metered, gated transaction — from one core.
JouleHook does not replace the host's agent. It is the host-binding layer for the
sibling standards: JCP decides and grants, Mesh meters at the model front
door, jcp-efficiency is the receipt. JouleHook adds the seam contract and the
per-host conformance profiles that make them portable.
- Licence: Apache-2.0 for reference code (
LICENSE); CC-BY-4.0 for specification text (LICENSE-SPEC). - Status: v0.1.0 draft. Specification published (
SPEC.md); reference workspacejoulehook-rs/ships the gate binding (joulehook-hook) for theclaude-codeandhooks-familyprofiles — one PreToolUse hook that gates by JCP capability (decided outside the model, every decision sealed) and speaks the native decision format of Claude Code, Cursor, Windsurf, and Antigravity behind a--hostselector;joulehook-statusline, the in-editor joule meter (honestestfrom Claude Code's own tokens); andjoulehook-mcp, an MCP server that surfaces the sealed decisions as verified, human-readable resources in any MCP-speaking editor. Theclaude-codefullprofile installs today (joulehook-rs/INSTALL.md); the enforcing Mesh-proxy meter + a site land next.
The three seams
| Seam | Host mechanisms | What rides it |
|---|---|---|
| Gate | PreToolUse-hook family; ACP request_permission |
allow / deny / rewrite, decided outside the model, sealed |
| Meter | base_url proxy; OTel gen_ai.* |
tokens→joules, honest provenance, enforceable joule ceiling |
| Receipt | MCP resources; status line; panel | hash-chained, verifiable efficiency/1 + decision receipts |
Conformance profiles
claude-code (reference) · hooks-family (Cursor / Windsurf / Antigravity) · acp
(Zed / JetBrains / Neovim) · opencode · cline · vscode (Open-VSX) · meter-only
(Continue, any OTel host). See SPEC.md §6.
The honest limit
A host-side adapter alone cannot meter the host agent's own tokens; a complete energy
figure requires the Meter seam (proxy or OTel). An adapter that claims whole-session
energy without a bound meter is non-conformant (SPEC.md §7).
What's here
joulehook/
├── SPEC.md — the normative standard (§1–§9)
├── README.md — this file
├── joulehook-rs/ — reference implementation
│ └── crates/
│ ├── joulehook-hook — ✅ the PreToolUse gate CLI for the hook family
│ │ (Claude Code / Cursor / Windsurf / Antigravity via --host;
│ │ wraps jcp-bridge::LocalToolGate for the JCP check + seal)
│ │ + joulehook-statusline — ✅ the in-editor joule meter
│ ├── joulehook-mcp — ✅ the receipt-surface MCP server (sealed decisions →
│ │ verified MCP resources, any MCP-speaking editor)
│ ├── joulehook-meter — ✅ the meter-seam binding: inject a grant's budget as
│ │ the Mesh joule ceiling (enforce) + bridge Mesh's
│ │ usage.energy into an efficiency/1 receipt
│ └── joulehook-proxy — ✅ the deployable meter: an OpenAI-shaped base_url
│ proxy that caps + seals every model call
└── conformance/v1/ — ✅ reproducible vectors: per-host decision encodings
(decisions.json) + a byte-deterministic sealed deny receipt
(deny-receipt.hex + receipt.json, verified-decode)
JouleHook is one of the Transaction Science open standards. Transaction Science is the steward, not the proprietor.