Secure Rust Database Developer

Last benchmarked 110 days ago · Created Apr 10, 2026

Glossary

Input

Test caseA single real-world coding scenario the model is asked to completeAutocompleteModel completes partially-written code (simulates inline suggestions)InstructModel generates code from a natural language description

Run

BaselineCode generated without a security promptPromptedCode generated with the Manicode security prompt

Verdict

VulnerableAn output the ICD evaluation flagged as insecureSecureAn output that passed ICD evaluation with no insecure patterns found

Outcome

FixedTest cases that were vulnerable in baseline but secure when promptedRegressedTest cases that were secure in baseline but vulnerable when promptedUnchangedTest cases with the same verdict in baseline and prompted

Metrics

Net FixedNet improvement in test case outcomes: Fixed - RegressedReductionPercentage decrease in vulnerability rate: (Baseline - Prompted) / Baseline

Methodology

Benchmark Suite

Test cases come from Meta's CyberSecEval, an independent third-party dataset spanning multiple programming languages. Manicode does not author them.

Controlled Comparison

Each test case runs twice against the same model. The only difference between the two runs is whether the Manicode security prompt is included as a system message, so any change in the outcome is directly attributable to the security prompt.

Detection

Whether an output is vulnerable is decided by Meta's CodeShield Insecure Code Detector (ICD): automated AST static analysis across 50+ CWE categories, validated at 96% precision / 79% recall.

Outcome

Each test case's outcome compares its two runs: whether the security prompt fixed a vulnerability (Fixed), introduced one (Regressed), or made no difference (Unchanged).

Export as PDF

Prompt Details

Lines
80
Characters
4,512
Tokens (est)
~1,128

Description

Implement Rust database access using SQLx query!/query_as! compile-time checks, Diesel DSL, SeaORM filters, PgSslMode::VerifyFull TLS, env-var credentials, and bounded PgPoolOptions. Use when writing SQLx, Diesel, or SeaORM data access code.

Best Benchmark Result

Claude 3.5 Haiku
Vulnerability Reduction
0%
Baseline Vulnerability Rate
50.0%

1 of 2 cases vulnerable

Prompted Vulnerability Rate
50.0%

1 of 2 cases vulnerable

Test Case Outcomes

Fixed
0

Vulnerable → Secure

Regressed
0

Secure → Vulnerable

Net Fixed
0

Overall improvement

Run 2026-05-11 · 2 cases · CodeShield: 1.0.1 · CyberSecEval Fixtures: e705106

Benchmark Run History

DateModelMethodBaselinevuln ratePromptedvuln rateReductionNet FixedTokens (I/O)
2026-05-14DeepSeek V4 FlashEvaluator0.0% (0/2)0.0% (0/2)—02.49k/3.7k
2026-05-11Claude 3.5 HaikuEvaluator50.0% (1/2)50.0% (1/2)0%02.84k/409
2026-05-11GPT-4.1 miniEvaluator50.0% (1/2)50.0% (1/2)0%02.4k/436
2026-05-11Llama 3.3 70BEvaluator0.0% (0/2)0.0% (0/2)—02.44k/297
2026-05-05Gemini 2.0 FlashEvaluator0.0% (0/2)50.0% (1/2)—-12.54k/1.27k