Skip to content

CHANGELOG.md

Latest commit

 

History

History
98 lines (76 loc) · 8.06 KB

CHANGELOG.md

File metadata and controls

98 lines (76 loc) · 8.06 KB

Changelog

All notable changes to CommitBee are documented here.

v0.5.0 — Beyond the Diff (current)

Semantic Analysis

  • Full signature extraction — The LLM sees pub fn connect(host: &str, timeout: Duration) -> Result<Connection>, not just "Function connect." Two-strategy body detection: child_by_field_name("body") primary, BODY_NODE_KINDS fallback. Works across all 10 languages.
  • Signature diffs for modified symbols — When a function signature changes, the prompt shows [~] old_sig → new_sig.
  • Cross-file connection detection — Detects when a changed file calls a symbol defined in another changed file. Shown as CONNECTIONS: validator calls parse() — both changed.
  • Semantic change classification — Modified symbols classified as whitespace-only or semantic via character-stream comparison. Formatting-only changes auto-detected as style.
  • Dual old/new line tracking — Correctly handles symbols shifting positions between HEAD and staged.
  • Token budget rebalance — Symbol section gets 30% of budget (up from 20%) when signatures present.

Security

  • Block project config URL overrides — .commitbee.toml can no longer redirect openai_base_url, anthropic_base_url, or ollama_host to prevent SSRF/exfiltration of API keys and staged code.
  • Cap streaming line_buffer — All 3 LLM providers cap line_buffer at 1 MB to prevent unbounded memory growth from malicious servers.
  • Strip URLs from error messages — reqwest::Error display uses without_url() to prevent leaking configured base URLs.
  • Broadened OpenAI secret pattern — Detects sk-proj- and sk-svcacct- prefixed keys alongside legacy sk- format.
  • Replaced Box::leak with Cow — Custom secret pattern names use Cow<'static, str> instead of leaked heap allocations.

Prompt Quality

  • Fixed breaking change subject budget — Subject character budget now accounts for ! suffix, preventing guaranteed validator rejection on breaking changes.
  • Omit empty EVIDENCE section — Saves ~200 chars when all flags are at default (most changes).
  • Symbol marker legend — SYSTEM_PROMPT now explains [+] added, [-] removed, [~] modified.
  • Removed duplicate JSON schema — System prompt no longer includes a competing schema template.
  • Replaced emoji with text — ⚠ replaced with WARNING: for better small-model tokenization.
  • Enhanced Python queries — Tree-sitter now captures decorated functions and classes.

Testing & Evaluation

  • Evaluation harness — 36 fixtures covering all 11 commit types, AST features, and edge cases. Per-type accuracy reporting with EvalSummary.
  • 15+ new unit tests — Coverage for detect_primary_change, detect_metadata_breaking, detect_bug_evidence (all 7 patterns), Deleted/Renamed status, signature edge cases, connection content assertions.
  • 5 fuzz targets — fuzz_sanitizer, fuzz_safety, fuzz_diff_parser, fuzz_signature, fuzz_classify_span.
  • 367 tests total (up from 308 at v0.4.0).

API

  • Demoted internal types — SymbolChangeType, GitService, Progress changed from pub to pub(crate).
  • Added #[non_exhaustive] to SymbolChangeType for future-safe extension.

v0.4.0 — See Everything

  • 10-language tree-sitter support — Added Java, C, C++, Ruby, and C# to the existing Rust, TypeScript, JavaScript, Python, and Go. All languages are individually feature-gated and enabled by default. Disable any with --no-default-features + selective --features lang-rust,lang-go,....
  • Custom prompt templates — User-defined templates with {{diff}}, {{symbols}}, {{files}}, {{type}}, {{scope}} variables via template_path config.
  • Multi-language commit messages — Generate messages in any language with --locale flag or locale config (e.g., --locale de for German).
  • Commit history style learning — Learns from recent commit history to match your project's style (learn_from_history, history_sample_size config).
  • Rename detection — Detects file renames with similarity percentage via git diff --find-renames, displayed as old → new (N% similar) in prompts and split suggestions. Configurable threshold (default 70%, set to 0 to disable).
  • Expanded secret scanning — 24 built-in patterns across 13 categories (cloud providers, AI/ML, source control, communication, payment, database, cryptographic, generic). Pluggable engine: add custom regex patterns or disable built-ins by name via config.
  • Progress indicators — Contextual indicatif spinners during pipeline phases (analyzing, scanning, generating). Auto-suppressed in non-TTY environments (git hooks, pipes).
  • Evaluation harness — cargo test --features eval for structured LLM output quality benchmarking.
  • Fuzz testing — cargo-fuzz targets for sanitizer and diff parser robustness.
  • Exclude files — --exclude <GLOB> flag (repeatable) and exclude_patterns config option. Glob patterns filter files from analysis (e.g., *.lock, **/*.generated.*, vendor/**). CLI patterns additive with config.
  • Copy to clipboard — --clipboard flag copies the generated message to the system clipboard and prints to stdout, skipping commit confirmation.

v0.3.1 — Trust, but Verify

  • Multi-pass corrective retry — Validator checks LLM output against 7 rules and retries up to 3 times with targeted correction instructions
  • Subject length enforcement — Rejects subjects exceeding 72-char first line with a clear error instead of silent truncation
  • Stronger prompt budget — Character limit embedded directly in JSON template, "HARD LIMIT" phrasing for better small-model compliance
  • Default model: qwen3.5:4b — Smaller (3.4GB), no thinking overhead, clean JSON output out of the box
  • Configurable thinking mode — think config option for Ollama models that support reasoning separation

v0.3.0 — Read Between the Lines

  • Diff-shape fingerprinting + Jaccard clustering — Splitter groups files by change shape and content vocabulary, not just directory
  • Evidence-based type inference — Constraint rules from code analysis drive commit type selection (bug evidence → fix, mechanical → style, dependency-only → chore)
  • Robust LLM output parsing — Sanitizer handles <think>/<thought> blocks, conversational preambles, noisy JSON extraction
  • Metadata-aware breaking change detection — Detects MSRV bumps, engines.node, requires-python changes
  • Symbol tri-state tracking — Added/removed/modified-signature differentiation in tree-sitter analysis
  • Primary change detection — Identifies the single most significant change for subject anchoring
  • Post-generation validation — Subject specificity validator ensures concrete entity naming
  • NUL-delimited git parsing — Safe handling of paths with special characters
  • Parallel tree-sitter parsing — rayon for CPU-bound parsing, tokio JoinSet for concurrent git fetching
  • Anti-hallucination prompt engineering — EVIDENCE/CONSTRAINTS sections, negative examples, anti-copy rules

v0.2.0 — Commit, Don't Think

  • Cloud providers — OpenAI-compatible and Anthropic streaming support
  • Commit splitting — Automatic detection and splitting of multi-concern staged changes
  • Git hook integration — commitbee hook install/uninstall/status
  • Shell completions — bash, zsh, fish, powershell via clap_complete
  • Rich error diagnostics — miette for actionable error messages
  • Multiple message generation — --generate N with interactive candidate selection
  • Hierarchical config — figment-based layering (CLI > Env > File > Defaults)
  • Structured logging — tracing with COMMITBEE_LOG env filter
  • Doctor command — commitbee doctor for connectivity and config checks
  • Secure key storage — OS keychain via keyring (optional feature)
  • Body line wrapping — Commit body text wrapped at 72 characters