Static reasoning can't prove runtime behavior: benchmark of AI PR reviewers
Comments on “Static reasoning can't prove runtime behavior: benchmark of AI PR reviewers”
3 judges. Scored by all three judges. Only the day’s top stories get the claim check. How far stories are checked
Heal, a runtime-testing agent, builds a sandbox around an application and drives it through its real entry points, such as UI, API, CLI, and webhooks, to reproduce defects with failing tests.
Why it ranked
No. 131 of 155 with a score of 0.56. The panel rated it highest on new information (0.85) and reader interest (0.75), lowest on significance (0.57). Overclaim counted against it (0.51). The judges disagreed most on promotional (spread 0.46), so the story is marked contested.
- AI judgeJudge 2claude-haiku-5-5
- Vendor benchmark of an AI runtime-testing tool vs PR review bots; useful framing and data, but self-promotional, with unverified claims and limited sample size.
- AI judgeJudge 3gpt-5.6-luna
- A detailed, data-backed benchmark shows runtime validation finding defects missed by PR bots, with methodology, examples, limitations, and actionable implications. Confidence is reduced by small samples, vendor self-interest, unvalidated review findings, and absolute framing.
Each judge wrote its line after answering the rubric, with the site and the author still hidden. Every answer behind this is in the table under Score.
Score
0.56total under rubric v2.2Contestedjudges differ by 0.44 on substantiated, 0.46 on promotional
- Significance0.57
- Reader interest0.75
- Audience breadth0.64
- Practical relevance0.73
- Substantiated0.69
- Technical depth0.71
- New information0.85
- Counts against: Promotional0.36
- Counts against: Overclaim0.51
Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).
| Question | Judge 1 | Judge 2 | Judge 3 | Mean | Spread | Weight | Contribution (weight times mean) |
|---|---|---|---|---|---|---|---|
| Significance | 0.65 | 0.35 | 0.72 | 0.57 | 0.37 | +0.24 | +0.138 |
| Reader interest | 0.75 | 0.60 | 0.90 | 0.75 | 0.30 | +0.26 | +0.195 |
| Audience breadth | 0.74 | 0.50 | 0.68 | 0.64 | 0.24 | +0.13 | +0.083 |
| Practical relevance | 0.76 | 0.60 | 0.82 | 0.73 | 0.22 | +0.22 | +0.160 |
| Substantiated | 0.89 | 0.45 | 0.72 | 0.69 | 0.44 , above the contested threshold | +0.05 | +0.034 |
| Technical depth | 0.80 | 0.55 | 0.77 | 0.71 | 0.25 | +0.05 | +0.035 |
| New information | 0.90 | 0.70 | 0.94 | 0.85 | 0.24 | +0.05 | +0.042 |
| Promotional | 0.14 | 0.60 | 0.34 | 0.36 | 0.46 , above the contested threshold | −0.19 | −0.068 |
| Overclaim | 0.36 | 0.55 | 0.62 | 0.51 | 0.26 | −0.11 | −0.056 |
| Sum of contributions | +0.564 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.564 | ||||||
Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.
The badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.
<a href="https://rubric.hrih.si/editions/2026-10-28/131/"><img src="https://rubric.hrih.si/badge/2026-10-28/131.svg" alt="Ranked #131 on Rubric, 28 Oct 2026" width="252" height="28"></a>Also covered by
No other coverage was clustered with this story.
Elsewhere
Elsewhere: HN 6 pts, not on front page
Recorded for context only, checked 28 Oct 2026, 06:47 UTC. It played no part in the ranking.