Skip to content

Commit e643524

Browse files
release: v0.4.0 - AI enhancements and Neo4j knowledge graph
New features: - Advanced prompt engineering with chain-of-thought reasoning - Neo4j knowledge graph for vulnerability similarity search (optional) - VulnCategory-specific few-shot examples - Optional feature flags (neo4j, redis) See CHANGELOG.md for full details. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
1 parent da2b098 commit e643524

2 files changed

Lines changed: 40 additions & 1 deletion

File tree

CHANGELOG.md

Lines changed: 39 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -5,6 +5,45 @@ All notable changes to MCPLint will be documented in this file.
55
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
66
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
77

8+
## [0.4.0] - 2025-12-14
9+
10+
### Added
11+
12+
- **Advanced Prompt Engineering** (`src/ai/prompt_templates.rs`)
13+
- `VulnCategory` enum for 8 vulnerability categories (Injection, Authentication, Cryptographic, etc.)
14+
- `FewShotExample` struct with category-specific examples
15+
- `AdvancedPromptBuilder` with chain-of-thought reasoning
16+
- Confidence scoring in AI explanations
17+
- Category-specific system prompts for better accuracy
18+
19+
- **Neo4j Knowledge Graph Integration** (optional `--features neo4j`)
20+
- `SecurityKnowledgeGraph` for storing and querying vulnerability findings
21+
- Vector similarity search using cosine distance
22+
- `VoyageEmbedder` for code-optimized embeddings (voyage-code-2, 1536 dimensions)
23+
- CWE/CVE knowledge retrieval
24+
- Cross-server vulnerability pattern detection
25+
26+
- **Optional Feature Flags**
27+
- `neo4j` - Neo4j graph database for vulnerability knowledge base
28+
- `redis` - Redis distributed cache backend
29+
30+
### Changed
31+
32+
- AI providers now support advanced prompts with `use_advanced_prompts` flag
33+
- Ollama provider defaults to simplified prompts for better local model performance
34+
35+
### Fixed
36+
37+
- Clippy lints for CI compatibility (needless borrows, single char push, duplicated cfg attributes)
38+
- Fuzzer session test overflow on Windows
39+
- ExplainEngine integration test flakiness with retry logic
40+
41+
### Tests
42+
43+
- Added 93+ new tests across modules
44+
- Total test count: 3,066 passing tests
45+
- Neo4j integration tests (require live connection)
46+
847
## [0.3.1] - 2025-12-13
948

1049
### Added

Cargo.toml

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
[package]
22
name = "mcplint"
3-
version = "0.3.0"
3+
version = "0.4.0"
44
edition = "2021"
55
authors = ["Russ Smith <quanticsoul4772@users.noreply.github.com>"]
66
description = "MCP Server Testing, Fuzzing, and Security Scanning Platform"

0 commit comments

Comments
 (0)