The systems language built for coding agents, and readable by humans.
Meaning in. Evidence-bound output.
Source remains a clear human projection. Coding agents work against typed meaning, stable identity, bounded context, and revision-bound changes. At the audited snapshot, all 49 full-product requirements are Partial, and none is Implemented.
revision sha256:<program-state>
query app.main --depth 1
context typed · bounded · stable-id
patch expected_revision == current
verify fail_closed
target native | browser/wasmRepository status at the audited snapshot
| Repository snapshot | c16348f · 2026-08-29. Documentation passed, but the overall workflow failed, so this head is not labelled verified. |
|---|---|
| Full product status | 49 Partial · 0 Implemented · 0 Missing |
| Graph and project contract | Graph ≤ v24 · Project v1 baseline · Project v8, v9, v10 developer preview |
| Promotion baseline | No exact passing promotion commit or workflow run is asserted for this snapshot. |
| Authored developer preview | Owned-data, record, and UTF-8 project profiles, package analysis, borrowing additions, Project Agent Transport, and Revision Store are present in source but unpublished or unpromoted. |
What is Semaprax?
Semaprax is an experimental agent-native systems programming language built around feature-selected semantic graph schemas through v24. Its v0.2 prototype includes bounded interpreter, C11/Clang, Core Wasm/Node, workspace, project, package, ownership, borrowing, transport, and revision-storage slices. Those slices are evidence of progress, not a completed product or stable general ABI.
One program, three working surfaces
Human source, agent-facing meaning, and machine output are derived from one versioned semantic state. Every arrow below is an evidence boundary, not a promise of universal language support.
- Human sourceReadable syntax remains the review and authorship surface.
- Semantic program graphStable IDs expose typed meaning, context, diagnostics, patch targets, ownership, cleanup, and borrowing facts through feature-selected schemas up to v24.
- Evidence-bound outputAdmitted subsets lower to interpreter, native, Core Wasm/Node, and bounded generated package lanes, each with an explicit evidence and publication state.
Semaprax semantic program graph: Semaprax parses human-readable source into validated HIR and a feature-selected semantic graph, preserving older graph bytes when newer schemas are not required. Graph v22 adds bounded owned-record and variant facts, v23 adds Shared Loan Plan facts, and v24 adds projected owned-byte-field borrowing. Project v1 remains the promoted baseline; Project v8-v10 package routes are authored developer previews, not supported public APIs.
The hypothesis, not the headline
Semaprax tests whether a programming system can reduce the context and repair ambiguity faced by coding agents. Lower model cost is a hypothesis. It is not yet supported by model-token, latency, answer-quality, or repository-scale evidence.
A research index with one owner per question
Each page answers one branded question so search engines and answer engines do not have to guess which route is authoritative.
Architecture
Stable IDs, graph context, and semantic patches
evidenceEvidence
What v0.2 implements and where it stops
benchmarksBenchmarks
Context measurements without cost promises
interoperabilityInteroperability
Target and ecosystem evidence matrix
roadmapRoadmap
Milestones, objectives, and open gaps
Capability evidence ledger
| Current status | Evidence state | Scope | Evidence |
|---|---|---|---|
| Feature-selected semantic graph | Partial | Deterministic legacy-schema preservation with owned-record and variant facts through v22, Shared Loan Plan facts in v23, and projected owned-byte-field borrowing in v24. | GitHub repository |
| Bounded agent context | Partial | Agent Context v1 plus additive v2 call traversal, with deterministic byte, node, depth, and frontier limits. | GitHub repository |
| Revision-bound semantic patches | Partial | Atomic Patch v1 and bounded Patch v2 operations, with exact revision checks, stale rejection, and selective semantic delta validation. | GitHub repository |
| Layered managed workspace changes | Partial | Workspace Operations v1 remains limited to declaration and import-alias renames. Separate change, replacement, structural, immutable-generation, and publication layers do not allow arbitrary tree or Git mutation. | GitHub repository |
| Evidence, publication, and revision storage | Partial | Review, impact, target, patch, and workspace evidence plus managed immutable generations and authenticated revision storage under bounded contracts. | GitHub repository |
| Project v1 baseline and v8-v10 previews | Partial | Project v8 adds Bytes, Option<Bytes>, and Result<Bytes, i64>; v9 adds flat owned records; v10 adds owned UTF-8, while preserving v1-v7 compatibility. | GitHub repository |
Questions, answered without the hype
Is Semaprax production-ready?
No. Semaprax is Apache-2.0 pre-alpha research. The repository calls itself an executable architectural seed, and the evidence ledger separates bounded implementation from general completion.
Is Semaprax only for AI agents?
No. The source projection is designed to stay readable by humans. Agent-native describes the stable semantic graph, bounded context, diagnostics, and patch interface available to coding agents.
Need this level of evidence-first engineering in your AI product?
See Wavect’s AI engineering work.