You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Godot GDScript (.gd) symbol outlining in read_slices (#2, thanks @star-dino)
Extracts class_name, nested inner class (with accurate depth), func / static func, single- and multiline signal, and named/anonymous single- and multiline enum blocks.
Added outputMode ("content" | "files_with_matches" | "count") to return only matching file paths or per-file match counts within the 8.8 KB inline budget.
Added fileType language/extension filtering (e.g., typescript, ts, python, gdscript) across text, regex, and fuzzy queries.
Added multiline: true cross-line text/regex matching across both rg and Node.js fallback (with automatic retry without strict u mode for ripgrep-valid identity escapes like \: / \#).
Added type: "fuzzy" case-insensitive multi-token path lookup (src/utils/fuzzy-match.ts) with Windows backslash normalization and contiguous-substring-first ranking.
Added sortBy: "mtime" for glob queries to list recently modified files first.
Documentation
Added sanitized 10-day real-world latency, token-quota, and upload-bandwidth benchmarks, an explicit non-invasive UI/diff guarantee, and a concise FAQ to README.md and README.zh-CN.md.