AgentsBestFriend

Atomically write new file contents to disk. WRITE TOOL — disabled unless ABF_ENABLE_WRITES=1 is set. Requires expected_old_hash (sha256 of the file currently on disk) to prevent overwriting concurrent changes. For new files pass expected_old_hash="" (empty string) and the file must not yet exist. AL…

Compute the transitive blast radius of a file (or symbol's defining file): which files would be affected if it changes. BFS over the reverse-import graph with depth control and a heuristic break-risk score.

Build a call graph around a function or method (TypeScript/JavaScript only, ts-morph). Modes: - "callers": who calls the symbol (transitive, by depth) - "callees": what the symbol calls (transitive, by depth) - "both": combined view Use this to answer "what happens when this function runs?" or "who …

Smart file chunking by symbol boundaries. Returns actual source code. Use EXACTLY ONE of these modes: - symbol: pass a symbol name to get its full source code directly - chunk_index: pass a 0-based chunk index to get that chunk's code - (neither): returns a chunk overview listing — use this first to…

Bundle multi-file context around an entry point in a single call. Traverses the import graph from the entry file, collecting symbols and optionally source code from all connected files. Replaces multiple read_file + abf_symbols + abf_dependencies calls with one compact result. Modes: - "signatures"…

Detect codebase conventions by analyzing file names, folder structure, design patterns, and config files. Returns detected conventions with confidence scores and examples. Purely heuristic — no LLM required. Useful to understand a project's style before making changes.

Jump to a symbol's definition in TypeScript/JavaScript code. Provide either (line + column) for a position-based lookup or (symbol) to find a named identifier in the file first. Returns absolute file path, line range, and a short source preview — replaces "search for the definition" round-trips.

Show imports and reverse dependencies (imported_by) for a file.

Run TypeScript diagnostics on a single file or every git-tracked TS/JS file. Surfaces type errors, missing imports, and other compile-time issues without running a full build. Use this AFTER editing TS/JS code to verify correctness before declaring a task done.

Search across LLM-generated file summaries using full-text search. Returns files whose summaries match the query, ranked by relevance (BM25). Useful when exploring a codebase by concept rather than exact code text. Default mode is "or" — matches files containing ANY of the query terms, ranked by how…

Query git history, blame, and diff for the project.

Get the inferred type and JSDoc for an identifier in TypeScript/JavaScript code. Equivalent to a "hover tooltip" in an IDE: returns the symbol's resolved type signature and any documentation comments. Use this to verify an API's exact shape before calling it.

AST-aware impact analysis for TS/JS — eliminates string/comment false positives. For non-TS/JS files, falls back to enriched ripgrep with comment filtering. Use this BEFORE renaming or removing a symbol when accuracy matters.

Find all files and lines that reference a given symbol name. Useful for change impact analysis.

Manage the file index: check status, trigger rebuild, incremental update, or generate LLM summaries. Rebuild/update will auto-generate summaries when Ollama is available.

Agent notepad — persist notes across sessions in a local SQLite database (.abf/notes.db). Supports save, get, list, search (full-text via FTS5), update, and delete actions. Use this to store task context, decisions, TODOs, or anything that should survive between conversations. Notes are project-scop…

Health check tool. Returns server status and project root. Use this to verify the ABF MCP server is running correctly.

Preview a planned file change WITHOUT writing to disk. Returns a structured diff plus symbol-level deltas, exported-API impact warnings, and a quick external-usage probe for removed/renamed exports. Use BEFORE editing to validate scope and risk.

Get a comprehensive overview of the current project. Returns detected tech stack, frameworks, entry points, directory structure with purposes, language distribution, config files, and architectural patterns. Very token-efficient way to orient yourself in a new codebase. No index required.

Generate a SAFE, ORDERED edit plan for a refactoring intent — read-only. Does NOT modify files. Currently supports: - rename: rename a symbol across the repo (best-effort, name-based) Other intents (move, extract, split) return a structured guidance plan.

Find tests that likely cover a given file or symbol. Heuristic ranking based on: file name match, test-file imports of the source, and direct mention of the symbol name in test code.

Run multiple search queries (mix of exact, keyword, semantic) in ONE call and merge results by file with weighted ranking. Reduces round-trips and combines complementary signals. Each sub-query contributes a normalized score; results are sorted by aggregated score.

Search code in the project using multiple modes: - "exact": Fast ripgrep-based search for exact strings or regex patterns. Returns matching lines with file path, line number, and surrounding context. - "keyword": Exploratory search — splits the query into keywords, scores every file by keyword densi…

Get the symbol outline (functions, classes, methods, types) of a file.