Skip to content
CAI
Software that uses CAICheck a score

cure-lab/ArbiterOS

32.2

Weak · 17 November 2026

360.5k

lines of production code

TypeScript

with Python

1

measurement over time

CAI band scale
CAI lens gauges

What this system is

Behavioural changes

1 commit (0 fixes) modifying (repo-wide)

A change to existing behaviour in (repo-wide) — 1 commit, 2693 files: “Add langfuse”.

(repo-wide) · medium confidence · unverified

17 commits (0 fixes) modifying ArbiterOS-Kernel/arbiteros\_kernel/policy

A change to existing behaviour in ArbiterOS-Kernel/arbiteros\_kernel/policy — 17 commits, 18 files: “add openclaw\_policy”; “Refine custom policy metadata flow”; “Add AlignmentSentinelPolicy to defaults”; and 12 more.

_ArbiterOS-Kernel/arbiteros\kernel/policy · medium confidence · unverified

2 commits (0 fixes) modifying ArbiterOS-Kernel/redteam

A change to existing behaviour in ArbiterOS-Kernel/redteam — 2 commits, 22 files: “add openclaw\_policy”; “Add files via upload”.

ArbiterOS-Kernel/redteam · medium confidence · unverified

2 commits (0 fixes) modifying scripts

A change to existing behaviour in scripts — 2 commits, 2 files: “add submodules for kernel and langfuse”; “script for init stack.env file”.

scripts · medium confidence · unverified

3 commits (0 fixes) modifying ArbiterOS-Kernel/arbiteros\_kernel/instruction\_parsing/shell\_parsers

A change to existing behaviour in ArbiterOS-Kernel/arbiteros\_kernel/instruction\_parsing/shell\_parsers — 3 commits, 4 files: “refactor: introduce tree-sitter-based bash parser as shell\_parsers/bash.py”; “refactor: restructure instruction\_parsing for platform-aware Windows/Linux support”; “refactor: decouple shell parsers from registries, add Darwin support”.

_ArbiterOS-Kernel/arbiteros\_kernel/instruction\_parsing/shell\parsers · medium confidence · unverified

3 commits (1 fix) modifying langfuse

A change to existing behaviour in langfuse — 3 commits (1 fix), 285 files: “Add langfuse”; “langfuse readme update”; “fix: enhance the readme”.

langfuse · medium confidence · unverified

4 commits (1 fix) modifying ArbiterOS-Kernel/arbiteros\_kernel/instruction\_parsing/registries

A change to existing behaviour in ArbiterOS-Kernel/arbiteros\_kernel/instruction\_parsing/registries — 4 commits (1 fix), 17 files: “feat: add macOS (Darwin) registry and unify user registry env var”; “fix: align YAML registries with agent instructions design spec”; “refactor: move skill\_trust to instruction\_parsing/helpers”; and 1 more.

_ArbiterOS-Kernel/arbiteros\_kernel/instruction\parsing/registries · medium confidence · unverified

44 commits (12 fixes) modifying arbiteros\_kernel

A change to existing behaviour in arbiteros\_kernel — 44 commits (12 fixs), 1 file: “fix: Improve tool error detection heuristics”; “refactor: restructure instruction\_parsing module with builder, types, and tool parsers”; “Update litellm\_callback.py”; and 40 more.

_arbiteros\kernel · medium confidence · unverified

49 commits (2 fixes) modifying ArbiterOS-Kernel/arbiteros\_kernel

A change to existing behaviour in ArbiterOS-Kernel/arbiteros\_kernel — 49 commits (2 fixs), 12 files: “Add role\_policy\_sets.json for role management”; “add openclaw\_policy”; “recover to 5 days ago”; and 38 more.

_ArbiterOS-Kernel/arbiteros\kernel · medium confidence · unverified

5 commits (0 fixes) modifying ArbiterOS-Kernel/arbiteros\_kernel/instruction\_parsing

A change to existing behaviour in ArbiterOS-Kernel/arbiteros\_kernel/instruction\_parsing — 5 commits, 2 files: “Add 'codex' to valid tool parsers”; “Add tool agent configuration management”; “support Hermes browser tools”; and 2 more.

_ArbiterOS-Kernel/arbiteros\_kernel/instruction\parsing · medium confidence · unverified

58 commits (4 fixes) modifying (repo-wide)

A change to existing behaviour in (repo-wide) — 58 commits (4 fixs), 8 files: “Add langfuse”; “support windows installation”; “update support long-term memory protect”; and 48 more.

(repo-wide) · low confidence · unverified

7 commits (0 fixes) modifying ArbiterOS-Kernel/arbiteros\_kernel/instruction\_parsing/tool\_parsers

A change to existing behaviour in ArbiterOS-Kernel/arbiteros\_kernel/instruction\_parsing/tool\_parsers — 7 commits, 3 files: “add instruction parsing dsl engine”; “add hermes registry”; “Clean up tool parser initialization”; and 4 more.

_ArbiterOS-Kernel/arbiteros\_kernel/instruction\_parsing/tool\parsers · medium confidence · unverified

Test coverage

2 commits adding/updating tests in ArbiterOS-Kernel/tests; 8 commits adding/updating tests in ArbiterOS-Kernel/tests/instruction\_parsing.

Dependencies

8 commits updating dependencies (11 manifests)

A dependency / build maintenance change in (dependencies) — 8 commits, 11 files: “Add langfuse”; “chore: configure ty to use project venv for module resolution”; “feat: add tree-sitter and tree-sitter-bash dependencies”; and 5 more.

(dependencies) · medium confidence · unverified

Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

How this codebase got here

Baseline

  • First survey — no prior run to compare against. CAI 32.2.

Lenses

  • Code Health 43.4
  • Architecture 38.7
  • Maturity 69.0
  • Readiness 26.6
  • Security 57.3
  • Domain Modelling 28.2
  • Accessibility 47.8
  • Performance 60.0

Changes since last survey

  • 300 commits — 272 feature/other, 28 fixes

By area

  • ArbiterOS-Kernel/arbiteros_kernel — 93 commits
  • arbiteros_kernel/instruction_parsing — 52 commits
  • (root) — 45 commits
  • (repo) — 23 commits
  • arbiteros_kernel/policy — 14 commits
  • arbiteros_kernel/litellm_callback.py — 11 commits
  • assets/docs — 10 commits
  • ArbiterOS-Kernel/litellm_config.yaml — 7 commits
  • ArbiterOS-Kernel/pyproject.toml — 6 commits
  • arbiteros_kernel/policy_check.py — 6 commits
  • ArbiterOS-Kernel/.claude — 4 commits
  • ArbiterOS-Kernel/docs — 3 commits
  • ArbiterOS-Kernel/redteam — 3 commits
  • ArbiterOS-Kernel/tests — 3 commits
  • redteam/case — 3 commits
  • arbiteros_kernel/policy.json — 2 commits
  • ArbiterOS-Kernel/README.md — 1 commit
  • ArbiterOS-Kernel/uv.lock — 1 commit
  • arbiteros_kernel/policy_test_harness.py — 1 commit
  • arbiteros_kernel/user_approval.py — 1 commit

Notable commits

  • fix: Fix policy_check.py: remove stale conflict markers from merge
  • fix: Fix subshell parenthesis stripping in exec pipeline classification
  • fix: Revert "Restore arbiteros-instruction-metadata-dsl-writer skill from feat/dsl-instr-parsing"
  • fix: add instruction logging at _commit; fix metadata for add_from_structured_output
  • fix: fix : only register absolute file path
  • fix: fix message bugs
  • fix: fix missing user registry
  • fix: fix pytest error
  • fix: fix some info in litellm_callback
  • fix: fix unused import
  • fix: fix wrong taint_status compute
  • fix: fix: Emit WARNING for inactivate error types
  • fix: fix: align YAML registries with agent instructions design spec
  • fix: fix: clean up logging in instruction_parsing
  • fix: fix: correct self-referencing $ref in tool_parsers schema
  • fix: fix: cron add/update should be parsed as SUBSCRIBE instruction.
  • fix: fix: enhance the readme
  • fix: fix: error-detection heuristic being interfered by taint watermark
  • fix: fix: failed unit tests
  • fix: fix: quote-aware shell pipeline splitting in _parse_exec
  • …and 280 more

Architecture

  • 0 containers · 1 bounded contexts · 0 dependency edges (baseline)

Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

Survey your own repository

cure-lab/ArbiterOS was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point a surveyor at a repository you know and see whether you agree with it.

About this page

  • The score is its most recent published measurement, taken on 17 November 2026 at a pinned commit. It is not a live figure and does not change until the project is measured again.
  • Measured at commit 78a8f98b1f1b4fdd2d875a058f52896cb588f8cf — the exact code this score is about.
  • Scored under rubric-2026.11.4 — the same rubric and the same method as every other entry in this index.
  • Measured by Canine Code Analyzer (cca.canine.dev) using cca measure-2026.11.85+preprod-eba6548ac50f.