Skip to main content
Version: 0.1.21

Why X07

X07 is an agent-first systems language: it’s designed so autonomous coding agents can generate, modify, test, and repair programs reliably.

The core bet is simple: agents converge faster and more safely when the language and toolchain are built around deterministic workflows, canonical representations, and machine-checkable boundaries.

The trust story (what makes agent code reviewable)​

  • Deterministic evaluation: keep correctness loops in solve-* worlds; use OS worlds only with explicit intent.
  • Record/replay: turn real OS interactions into deterministic cassettes you can re-run in fixture worlds.
  • Budgets: local budget.scope_v1 caps and arch-driven budget profiles prevent cost blowups from small agent edits.
  • Repair loops: the toolchain surface is JSON-first (diagnostics + patches + reports), so “lint → fix → re-run” is machine-drivable.
  • Review artifacts: semantic diff + trust report make changes auditable (world/capability deltas, budgets, nondeterminism flags).

Evidence pack (public, reproducible)​

1) Cross-language performance comparisons​

For runtime/compile time/binary size comparisons (X07 vs C vs Rust vs Go), use:

  • x07lang/x07-perf-compare (runs locally, verifies output equivalence)

2) Agent correctness benchmark harness (x07 bench)​

x07 bench evaluates patch submissions against versioned benchmark suites with deterministic artifacts and a machine-readable report:

  • Seed suite: labs/x07bench/suites/core_v1/ (expanded; recommended)
  • Docs: Benchmarks

3) Diagnostic catalog + quickfix coverage gate​

X07 tracks diagnostics as a catalog and enforces quickfix coverage as a CI gate:

  • Catalog tooling: x07 diag catalog, x07 diag check, x07 diag coverage
  • Rendered codes doc: Diagnostic codes

4) Agent-consumable spec + tool contracts​

The toolchain exposes stable, machine-readable surfaces for agents:

  • Schemas: spec/*.schema.json
  • Offline docs + agent portal endpoints: Agent contracts
  • Machine doc API: x07 doc --json ...

How to evaluate locally (one path)​

  1. Start from Agent quickstart.
  2. Run an end-to-end workflow on one of the reference projects under docs/examples/.
  3. Run x07 bench validate / x07 bench eval on labs/x07bench/suites/core_v1/.
  4. Run x07-perf-compare for cross-language perf + build size comparisons.