Catch Go Bugs That Fuzzing Misses with Gemini and a Constraint Solver
Comments on “Catch Go Bugs That Fuzzing Misses with Gemini and a Constraint Solver”
Claim-checked. Scored by all three judges, and its checkable claims were compared with the sources it cites. How far stories are checked
A Google engineer describes a Go linting workflow that pairs Gemini 4 Argon with the Z3 constraint solver to find runtime panics that survive Go's bounds-check elimination.
Why it ranked
No. 30 of 60 with a score of 0.62. The panel rated it highest on technical depth (0.93) and new information (0.90), lowest on audience breadth (0.46). Placed No. 30, No. 26 by score alone: the sixth Programming & dev tools story of the day (−0.060). The placement terms move stories; they are not part of the score.
- AI judgeJudge 2claude-haiku-5-5
- Concrete engineering write-up on Go bounds-check bugs using an LLM-driven Z3 pipeline; technically deep with reproducible examples, but a narrow audience and unverified claims about a solver-based workflow.
- AI judgeJudge 3gpt-5.6-luna
- A detailed, reproducible Go engineering workflow combines compiler disassembly, Gemini-generated Z3 constraints, and concrete fuzzing comparisons. It offers actionable testing guidance and reports real bugs, though its impact is mainly concentrated among Go developers.
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
0.62total under rubric v2.3
- Significance0.48
- Reader interest0.77
- Audience breadth0.46
- Practical relevance0.76
- Substantiated0.80
- Technical depth0.93
- New information0.90
- Counts against: Promotional0.11
- Counts against: Overclaim0.28
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) |
|---|---|---|---|---|---|---|---|
| SignificanceHow much does this change what people in tech know or can do? | 0.51 | 0.35 | 0.58 | 0.48 | 0.23 | +0.24 | +0.115 |
| Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on? | 0.74 | 0.70 | 0.86 | 0.77 | 0.16 | +0.26 | +0.199 |
| Audience breadthHow many people who build technology does it concern? | 0.61 | 0.40 | 0.38 | 0.46 | 0.23 | +0.13 | +0.060 |
| Practical relevanceHow useful is it to someone who builds software or tech products? | 0.85 | 0.60 | 0.82 | 0.76 | 0.26 | +0.22 | +0.167 |
| SubstantiatedHow well is it backed? Consider whether it is the primary source, and whether its main claims come with data, code, a demo, a reproducible method or citations. | 0.97 | 0.60 | 0.84 | 0.80 | 0.37 | +0.05 | +0.040 |
| Technical depthHow technically substantive is it? | 1.00 | 0.85 | 0.93 | 0.93 | 0.15 | +0.05 | +0.046 |
| New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)? | 0.86 | 0.90 | 0.94 | 0.90 | 0.08 | +0.05 | +0.045 |
| PromotionalIs it mainly marketing with little substance (a press release, ad or sponsored post with no real information)? A launch post that carries real news is not promotional. | 0.06 | 0.20 | 0.08 | 0.11 | 0.14 | −0.19 | −0.022 |
| OverclaimDo the headline or conclusions claim more than the body supports? | 0.28 | 0.45 | 0.12 | 0.28 | 0.33 | −0.11 | −0.031 |
| Sum of contributions | +0.620 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.620 | ||||||
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/2026-11-16/30/"><img src="https://rubric.hrih.si/badge/2026-11-16/30.svg" alt="Ranked #30 on Rubric, 16 Nov 2026" width="245" height="28"></a>Claim check
Claim check: 1 matches source · 6 no source given · 1 couldn’t check. Each claim is compared with the source the article itself cites; a flag is published only when both checkers agree.
No source givenNumber
The Gemini and Z3 workflow caught all three bugs planted in test parsers and turned up three latent bugs in skill-lint.
The article gives no link or citation for it.
The article says “That workflow caught all three bugs I planted in test parsers and turned up three latent bugs in skill-lint”
The source says No source to compare with.
Matches sourceGeneralization
Anthropic published a post showing that formal verification catches latent bugs in C and C++ by having frontier AI models write machine-checked Rocq proofs across OSS-Fuzz libraries.
The cited source says the same thing.
The article says “A recent post from Anthropic showed how formal verification catches latent bugs in C and C++ by having frontier AI models write machine-checked Rocq proofs across OSS-Fuzz libraries”
The source says “Over the span of a few weeks, Claude successfully patched the bugs it found and then proved that there are no crashes on any input (up to 64 MB) for…”
Source cited by the article: anthropic.com/research/formal-verification
Couldn’t checkCausal conclusion
Go's SSA prove pass eliminates most bounds checks before machine code generation by tracking slice lengths and loop variables.
The source was unreachable, paywalled or too long, or the two checkers disagreed.
The article says “Go's SSA prove pass already eliminates most of those checks by tracking slice lengths and loop variables”
The source says No passage could be compared.
Source cited by the article: github.com/golang/go/blob/go1.27.3/src/cmd/compile/internal/ssa/prove.go
No source givenNumber
Argon's translation took about a minute per function, while Z3 finished in milliseconds.
The article gives no link or citation for it.
The article says “Argon's translation dominated the runtime, at about a minute per function, while Z3 itself finished in milliseconds”
The source says No source to compare with.
No source givenNumber
Z3 returned a 59-byte counterexample for ParseTwoChunkFrame with hdrLen = 48 and bodyLen = 236, whose uint8 sum wraps from 284 to 28.
The article gives no link or citation for it.
The article says “Z3 returned a 59-byte counterexample with hdrLen = 48 and bodyLen = 236: their uint8 sum wraps from 284 down to 28”
The source says No source to compare with.
No source givenBenchmark result
Unseeded go test -fuzz missed the ParseTwoChunkFrame crash in one 3-second run and three 60-second runs, even though 19.4% of (hdrLen, bodyLen) pairs trigger it at 59 bytes.
The article gives no link or citation for it.
The article says “At 59 bytes, 19.4% of all possible (hdrLen, bodyLen) pairs trigger this crash. Yet a 3-second fuzz run missed it, and so did three 60-second runs”
The source says No source to compare with.
No source givenCausal conclusion
In a 10-second probe of 386,741 fuzz calls, 89% of inputs were views into a shared ~100 MB buffer and only 0.07% had cap == len, so the fuzzer rarely passes slices without spare capacity.
The article gives no link or citation for it.
The article says “In a 10-second probe of 386,741 fuzz calls, 89% of inputs were views into a shared ~100 MB buffer, and only 0.07% had cap == len”
The source says No source to compare with.
No source givenBenchmark result
After clipping capacity to length, the fuzzer crashed both ParseTwoChunkFrame and ExtractQuotedToken in all three runs, with median times of 0.28 s and 0.17 s.
The article gives no link or citation for it.
The article says “After that change, the fuzzer crashed both ParseTwoChunkFrame and ExtractQuotedToken in all three runs (median 0.28 s and 0.17 s”
The source says No source to compare with.
Also covered by
No other coverage was clustered with this story.
Elsewhere
Elsewhere: HN 2 pts, not on front page
Recorded for context only, checked 16 Nov 2026, 05:47 UTC. It played no part in the ranking.