Seven structural checks on a public repository. Scanned 2026-11-02 23:18 · 1,695 files
77.8/ 100 · risk level Medium
Strong delivery controls (CODEOWNERS, CI gate semantics, changelog) and excellent test reach. Two gaps: 50 business-logic files over 800 lines, and no cross-session memory arbitration. An agent editing this repo has good rails but a large blind spot in file size.
Scorecard
#
Check
Score
Weight
Verdict
C1
Scope Containment
7/10
18%
acceptable
C2
False-Green Detection
10/10
18%
pass
C3
Drift Containment
10/10
14%
pass
C4
Memory Integrity
4/10
10%
needs fixing
C5
Delivery Gate
10/10
16%
pass
C6
Complexity Control
1/10
12%
needs fixing
C7
Regression Surface
10/10
12%
pass
Evidence
C1 Scope Containment7/10
CI config: .github/workflows
CODEOWNERS present: .github/CODEOWNERS
Pre-commit hook: none
CI contains gate/guard/policy semantics: yes
C2 False-Green Detection10/10
Source files 1034 / test files 531 (test ratio 51.4%)
Constraint strength: 77 hard prohibitions found (MUST / NEVER / forbidden)
C4 Memory Integrity4/10
Memory/state files: none found
Project shows agent/LLM characteristics: yes
Note: agent project without cross-session memory files. Within a single session this is not a conflict; it becomes a gap the moment state is expected to persist
Conflict arbitration: none (no arbiter when old and new context disagree)
Risk: No state anchor across sessions. The same question gets contradictory answers on different days.
Three largest files: tests\test_run_state.py (12134 lines); tests\test_agent_runner.py (7638 lines); tests\sandbox\test_runtime.py (7403 lines)
Mean above 400 lines/file: granularity too coarse overall
Risk: Oversized files. The agent's context truncates mid-edit and the regression lands unseen.
C7 Regression Surface10/10
Dependency lock: uv.lock
Source packages 5, reachable by test imports 2 (coverage 40%); not covered: .agents, .github, docs
CI matrix across environments: yes
Priority stack
#
Fix first
Score
Why
1
Complexity Control
1/10
Oversized files. The agent's context truncates mid-edit and the regression lands unseen.
2
Memory Integrity
4/10
No state anchor across sessions. The same question gets contradictory answers on different days.
3
Scope Containment
7/10
below threshold
Want this run on your repository?
A 7-day, human-verified audit: scored report, priority stack, and a recorded walkthrough of your own code. Findings are checked by hand and anything that does not hold up is deleted.