tude-alt/testrun

• Repository: https://github.com/tude-alt/testrun
• Commit: 26e86c3f8732b70fe673456bf74108a716ad1863
• Generated: 2026-11-14T14:52:08+00:00
• Scorer 0.1.0, scoring prov-5+qual-2+sec-2, report schema v2

Scores

AI provenance

Weak signals consistent with AI-assisted development.

• Explicitly attributed share (lower bound): 0.0%

• Unattributed share in AI-consistent velocity bursts: 59.5%

• Estimated AI-generated share: 0.0%

• AI tooling configured in repo: claude-code, generic-agent

• Stylometry index: 0.15 per 1,000 lines (narrated steps 0, placeholders 0, assistant phrasing 0, emoji 31). Reported as a signal, not scored.

Evidence

• .github/workflows/claude.yml runs anthropics/claude-code-action (.github/work)
• 11,098 lines of first-party code in ~3.2h (3,465 lines/hour, 44 commit session) (c3bd3efc04e1)
• 2,840 lines of first-party code in ~1.0h (2,757 lines/hour, 2 commit session) (f213025f7e8c)
• 4,287 lines of first-party code in ~1.6h (2,647 lines/hour, 33 commit session) (e12135d87d0c)
• 6,861 lines of first-party code in ~2.7h (2,511 lines/hour, 30 commit session) (ea1b29a6cc8f)
• 6,050 lines of first-party code in ~2.4h (2,485 lines/hour, 8 commit session) (d30a9c4ea297)
• 2,859 lines of first-party code in ~1.4h (2,010 lines/hour, 1 commit session) (2b4690978e1f)
• 10,811 lines of first-party code in ~5.4h (1,993 lines/hour, 28 commit session) (820b1a701f90)
• 3,393 lines of first-party code in ~1.7h (1,970 lines/hour, 8 commit session) (6638350ac2bd)
• 6,915 lines of first-party code in ~4.0h (1,718 lines/hour, 8 commit session) (ff20e5e552ce)
• 9,665 lines of first-party code in ~5.6h (1,718 lines/hour, 33 commit session) (972e4d6212e5)
• AGENTS.md (generic-agent configuration) (AGENTS.md)
• 9,849 lines of first-party code in ~6.0h (1,640 lines/hour, 13 commit session) (a61c53d1c3ae)
• 6,736 lines of first-party code in ~4.3h (1,572 lines/hour, 7 commit session) (3e484edcc1a4)
• 2,547 lines of first-party code in ~1.6h (1,545 lines/hour, 2 commit session) (f852f8a53e53)

AI provenance describes how code was produced; it is not a quality judgment. Unless attribution is explicit, these are signals consistent with AI generation, not proof. Squash/rebase workflows, imported code, and rewritten history reduce accuracy.

Code quality

Code quality is adequate: test-to-source ratio 1.78, CI runs tests; 6% of function code in complex or long functions; 1% duplicated lines. Weakest area: type safety.

• Functions: 3,421; mean complexity 4.3, 90th percentile 9; 51 complex and 5 long functions
• Duplicated lines: 1.4% of 77,650 compared
• Tests: 182,025 test lines for 102,156 source lines (ratio 1.78); CI test commands: pytest, npm run test; 75% of recent source commits also change tests
• Docs missing or thin: contributing, changelog

Most complex functions:

• new_wave_summary in metaculus_bot/performance_analysis/round_outputs.py: complexity 37, 55 lines
• sweep_row in metaculus_bot/performance_analysis/clip_threshold_sweep.py: complexity 30, 45 lines
• _type_report in metaculus_bot/performance_analysis/clip_threshold_tables.py: complexity 29, 49 lines
• counts_payload in metaculus_bot/performance_analysis/round_outputs.py: complexity 28, 94 lines
• _ladder_at_stage in metaculus_bot/research/market_retrieval/rendering.py: complexity 28, 30 lines

Security

Security posture is poor: 156 known vulnerabilities in 17 dependencies; 10 high-severity SAST findings; 30 high-severity workflow issues. Weakest area: posture.

• Known vulnerabilities: 156 in 17 packages (critical 11, high 81, medium 54, low 9)

• Posture checks passed: lockfiles; missing: security policy, dependency updates, signed commits

Top findings

Static signals only: the scorer never runs the project's code. Secret findings are pattern matches that were not verified, and secret values are never stored.

Distribution signals

• readme-download-lure: gives an archive password

Signals consistent with a repository that exists to distribute a download. They are heuristics, not a verdict; release files were not downloaded or inspected.

Repository profile

• Source lines: 102,156; test lines: 182,025