graphifyy

• Cross-file call resolution for Go, Rust, Zig, PowerShell, and Elixir (raw_calls saved and resolved globally)
• Fix semantic/AST node ID mismatch: build.py now reconciles mismatched IDs so LLM-generated edges survive the merge
• skill.md: explicit node ID generation rules with example to reduce LLM ID drift

• Move llm.py out of graphify package into scripts/ so it is not shipped in the PyPI wheel

• Fix NameError: name 'sys' is not defined in graphify watch (v0.4.15 regression)
• Fix .mjs files detected but producing 0 nodes (missing _DISPATCH entry)
• Exclude llm.py from wheel

What's new

• VS Code Copilot Chat — graphify vscode install configures VS Code Copilot Chat with a Python-only skill (works on Windows PowerShell) and writes .github/copilot-instructions.md for always-on graph context (#206)
• Fix: OpenCode plugin path used backslashes on Windows causing duplicate opencode.json entries — fixed via .as_posix() (#378)
• Fix: Gemini CLI on Windows now installs skill to ~/.agents/skills/ (correct higher-priority path) instead of ~/.gemini/skills/ (#368)
• Fix: .mjs and .ejs files now recognised by the AST extractor as JavaScript (#365)
• Fix: god_nodes() return field renamed edges → degree for clarity
• Fix: graphify watch on macOS now uses PollingObserver by default to avoid missed events with FSEvents (#373)

Upgrade

pip install --upgrade graphifyy

What's new

Cross-file call resolution for all languages

AST extraction now emits cross-file calls edges for Swift, Go, Rust, Java, C#, Kotlin, Scala, Ruby, PHP, and every other supported language. Previously only Python had this. Unresolved call sites are saved per file and resolved against a global label map after all files are processed. Cross-file edges are marked INFERRED with confidence 0.8.

PHP static method calls and constant references

Two node types were missing from the PHP extractor. scoped_call_expression (static calls like Helper::format()) now emits calls edges. class_constant_access_expression (enum and constant references like Status::ACTIVE) now emits references_constant edges.

Wiki flag implemented

--wiki now actually runs. Step 6b added to the skill pipeline (before the cleanup step so community labels are still available). Calls to_wiki() and writes graphify-out/wiki/ with index.md and one article per community and god node.

Performance: betweenness centrality on large graphs

edge_betweenness_centrality now returns early for graphs over 5000 nodes. betweenness_centrality in suggest_questions uses k=100 approximate sampling for graphs over 1000 nodes. Eliminates hangs on large repos.

OpenCode plugin installed on both install paths

graphify install --platform opencode now also writes .opencode/plugins/graphify.js and registers it in opencode.json. Previously only graphify opencode install did this.

Cache root fix for subdirectory runs

extract() accepts an explicit cache_root parameter. When graphify narrows to a subdirectory on a large repo, the cache stays at ./graphify-out/cache/ instead of <subdir>/graphify-out/cache/.

Windows stability

os.replace in the cache writer falls back to shutil.copy2 on PermissionError (WinError 5). graphify update exits with code 1 when rebuild fails instead of silently returning. All generated config files now use graphify update . instead of a hardcoded python3 invocation.

Kiro package data

skill-kiro.md was missing from pyproject.toml package-data, causing graphify kiro install to fail on fresh pip installs. Fixed.

Upgrade

pip install -U graphifyy

What's new

Verilog / SystemVerilog support

• .v and .sv files now extracted via tree-sitter-verilog
• Extracts modules, functions, tasks, package imports, and instantiations

HiDPI hyperedge fix

• Hyperedges now render correctly on high-DPI displays (Retina, 4K)
• Fixed double-applied view transform in the afterDrawing canvas callback

Null label guards in MCP server

• Fixed crashes when nodes have null label or source_file fields in query_graph, get_neighbors, and get_node tools

AGENTS.md / GEMINI.md rebuild rule

• Replaced hardcoded python3 invocation with graphify update .
• Uses the correct interpreter (pipx, venv, system) and is properly scoped to the project subfolder

Upgrade

pip install -U graphifyy

What's new

Kiro IDE/CLI support (#319, #321)

graphify kiro install writes:

• .kiro/skills/graphify/SKILL.md — the /graphify slash command skill
• .kiro/steering/graphify.md — always-on steering with inclusion: always, injected into every Kiro conversation automatically

Portable cache (#311)

file_hash() now uses the path relative to the project root instead of the absolute path. Cache entries are now portable across machines, CI runners, and different checkout directories.

Note: this is a cache-breaking change. Delete graphify-out/cache/ once after upgrading — it will be rebuilt on the next run.

pip install --upgrade graphifyy
rm -rf graphify-out/cache/   # one-time after upgrading

Bug fixes

• graphify query crashes with ValueError on MultiGraph — G.edges[u, v] replaced with G[u][v] + MultiGraph guard (#305)
• AttributeError: 'NoneType' has no attribute 'lower' in serve.py when a node has a null source_file (#307)
• MCP server launched from a different directory now correctly derives the graphify-out base from the absolute path, not CWD (#309)
• .graphifyignore patterns from a parent directory now fire correctly when graphify is run on a subfolder (#303)

pip install --upgrade graphifyy

What's new

Dart/Flutter support

.dart files extracted via regex — classes, mixins, functions, imports. graphify now supports 23 languages.

Hermes Agent platform

graphify install --platform hermes writes the skill to ~/.hermes/skills/graphify/SKILL.md.

6 new terminal CLI commands

graphify path "AuthModule" "Database"   # shortest path between two nodes
graphify explain "SwinTransformer"      # plain-language node explanation
graphify add https://...                # fetch URL, save to ./raw, update graph
graphify watch ./src                    # rebuild on code changes, no LLM needed
graphify update ./src                   # re-extract code files only
graphify cluster-only .                 # rerun clustering on existing graph.json

PHP extractor improvements

Static property access, config() helper calls, service container bindings, event listener arrays.

Other fixes

• graspologic python_version < '3.13' marker (#290)
• matplotlib declared in [svg] extra (#288)
• norm_label in graph.json for diacritic-insensitive search (#293)
• prune_dangling_edges() utility (#294)
• cursor install crash fixed (#281)
• agents uninstall platform-aware (#276)
• Codex skill: wait → wait_agent (#273)
• Windows hook tests platform-aware (#279)
• Antigravity: YAML frontmatter injection + MCP config snippet (#268)

pip install --upgrade graphifyy

What's new

Bug fixes

• graphify install --platform cursor no longer crashes (#281)
• _agents_uninstall now only removes the OpenCode plugin for the opencode platform (#276)
• Removed misleading path guard comment in query --graph handler (#278)
• skill-codex.md: wait → wait_agent (correct Codex tool name) (#273)
• graspologic now has python_version < '3.13' env marker — fixes install on Python 3.13+ (#290)
• Windows hook tests now use platform-aware assertions instead of POSIX executable bit checks (#279)

New language support

• Dart/Flutter — .dart files extracted via regex (classes, mixins, functions, imports) (#292)

PHP extractor improvements

• Static property access (Foo::$bar) → uses_static_prop edges (#234)
• config() helper calls → uses_config edges (#236)
• Service container bindings (bind, singleton, scoped, instance) → bound_to edges (#238)
• Event listener arrays ($listen, $subscribe) → listened_by edges (#240)

Search improvements

• Diacritic-insensitive search — norm_label written at build time via Unicode NFKD normalization; query, get_node, shortest_path all benefit (#293)

New platform

• Hermes Agent — graphify hermes install writes skill to ~/.hermes/skills/graphify/SKILL.md + AGENTS.md (#251)

Other additions

• svg = ["matplotlib"] optional extra; matplotlib added to [all] (#288)
• prune_dangling_edges() utility in export.py — removes edges with missing source/target nodes (#294)
• Antigravity install: injects YAML frontmatter for native tool discovery, adds MCP navigation rule, prints MCP config snippet (#268)

433 tests passing.

Fix

Platform skill files (aider, codex, opencode, claw, droid, copilot, windows) no longer contain Claude-specific language. References to 'Claude' as the AI model are replaced with platform-agnostic wording so they work correctly on any agent. (#272)

PyPI: https://pypi.org/project/graphifyy/0.4.8/

Fixes

• Watch semantic edge loss — watch was silently dropping all semantic edges on every rebuild because graph.json uses the links key but code read edges. Now reads both (#269)
• OpenClaw install path — graphify claw install now correctly writes to .openclaw/skills/graphify/ (#208)
• Blade template support — .blade.php files now extracted: @include, <livewire:\> components, and wire:click bindings (#242)
• WSL/Linux MCP docs — added note covering graphifyy package name, venv setup, and full path in .mcp.json (#250)

PyPI: https://pypi.org/project/graphifyy/0.4.7/

What's new

Google Antigravity support

graphify now works with Google Antigravity — the agent-first IDE built on Gemini.

pip install --upgrade graphifyy
graphify antigravity install

This writes:

• .agent/rules/graphify.md — always-on rules (Antigravity reads these every session)
• .agent/workflows/graphify.md — registers /graphify as a slash command

Then open Antigravity and type /graphify .

---

Full changelog: https://github.com/safishamsi/graphify/blob/v4/CHANGELOG.md

PyPI: https://pypi.org/project/graphifyy/0.4.6/

---

If graphify has been useful, consider sponsoring — we're closing in on 100k PyPI downloads and every bit of support helps keep the project moving.

Bug fix for #201.

Fix

• #201 MCP server no longer crashes with ValidationError when an MCP client sends blank lines between JSON messages. A background thread now filters blank-only lines from stdin before the MCP stdio transport reads them.

Upgrade

pip install --upgrade graphifyy

Bug fixes for #249, #261, #266.

Fixes

• #261 graphify watch now preserves INFERRED/AMBIGUOUS edges (e.g. code↔doc rationale links) across AST rebuilds — previously any edge with one code-node endpoint was dropped, losing ~95 cross-type edges in mixed repos
• #249 Codex hook no longer emits permissionDecision:allow which codex-cli 0.120.0 rejects — switched to additionalContext (run graphify codex install to update your hook)
• #266 Common lockfiles (package-lock.json, yarn.lock, pnpm-lock.yaml, Cargo.lock, poetry.lock, go.sum, etc.) are now skipped during file detection — prevents token drain on large JS/Rust/Python projects

Upgrade

pip install --upgrade graphifyy

Bug fixes for issues #226, #244, #253, #254, #256.

Fixes

• #256 JS/TS relative imports (./foo) now resolve to full-path node IDs — previously all imports_from edges were silently dropped on large TypeScript codebases. Python relative imports fixed the same way.
• #253 graphify watch now merges fresh AST nodes with existing semantic nodes (docs/papers) from the last full run — previously every code save wiped the semantic layer.
• #244 Windows hooks now fall back to python if python3 is not on PATH; exits cleanly (no error) if neither has graphify installed.
• #226 surprising_connections / suggest_questions no longer crash with KeyError on stale _src/_tgt edge hints after a custom node merge or canonicalization pass.
• #254 .vue and .svelte files are now recognized as code and included in extraction.

Upgrade

pip install --upgrade graphifyy

Bug fixes — 8 issues addressed.

Bug fixes (our commits)

• #211 Same-basename files produced colliding node IDs — now uses full path
• #216 Edges using from/to keys were silently dropped — now normalized
• #217 Empty graphs caused ZeroDivisionError during visualization — guarded
• #222 Post-commit hook silently skipped .tsx, .jsx etc — removed stale CODE_EXTS filter

Community fixes (cherry-picked)

• #212 NetworkX ≤3.1 serialises edges as links — now accepted alongside edges
• #220 Version warning fired during install/uninstall and duplicated on shared paths — fixed
• #204 All file IO now uses encoding="utf-8" — prevents crashes on Windows CJK/emoji corpora; hook writes use newline="\n" to prevent CRLF shebang breakage
• #221 Obsidian: node labels ending in .md produced .md.md filenames; GRAPH_REPORT.md now links to community hub files so the vault stays in one connected component

Upgrade

pip install --upgrade graphifyy

Two bug fixes:

.graphifyignore not respected by collect_files() (#188) — collect_files() in extract.py was ignoring .graphifyignore patterns entirely. On projects with node_modules/ ignored, this caused 40k+ files to be scanned instead of ~200. Fixed by loading and applying ignore patterns the same way detect() does.

Semantic extraction silently dropped with Explore subagent (#195) — if Claude Code dispatched extraction subagents using subagent_type="Explore", chunk files were never written to disk (Explore is read-only), causing up to 2.4x fewer nodes extracted with no warning. skill.md now explicitly requires general-purpose subagent type and Step B3 warns when chunk files are missing instead of silently skipping.

Run pip install --upgrade graphifyy then graphify install to get both fixes.

Video and audio files are now first-class corpus inputs.

Drop .mp4, .mp3, .wav, .mov, .webm, .m4a, .ogg, .mkv, .avi, or .m4v files into any folder and graphify transcribes them locally with faster-whisper before extraction. Transcripts are cached in graphify-out/transcripts/ so re-runs skip already-transcribed files.

You can also add YouTube links directly:

/graphify add https://www.youtube.com/watch?v=...

yt-dlp downloads audio-only (no video stream, much smaller) and the same Whisper pipeline picks it up from there.

The Whisper prompt is built from your corpus automatically. The coding agent reads the top god nodes from your non-video files and writes a one-sentence domain hint for Whisper itself. No separate API call, no configuration needed.

To use:

pip install 'graphifyy[video]'

This installs faster-whisper and yt-dlp. The default Whisper model is base. Pass --whisper-model medium or --whisper-model large-v3 for higher accuracy.

Audio never leaves your machine. Transcription runs fully locally.

• Add: video and audio corpus support — drop .mp4, .mp3, .wav, .mov, .webm, .m4a, .ogg, .mkv, .avi, .m4v files into any corpus and graphify transcribes them with faster-whisper before extraction
• Add: YouTube and URL video download — pass a YouTube link (or any video URL) to /graphify add <url> and yt-dlp downloads audio-only, which is then transcribed and added to the corpus automatically
• Add: domain-aware Whisper prompts — god nodes from non-video files are used to build a one-sentence domain hint for Whisper via a cheap Haiku call, improving transcript accuracy on technical content
• Add: graphify-out/transcripts/ cache — transcripts are cached by filename so re-runs skip already-transcribed files; URLs cached by hash
• Requires: pip install 'graphifyy[video]' for faster-whisper + yt-dlp

Three bug fixes:

• Hook reinstall: Claude Code, Codex, and Gemini CLI hook installers now always remove and reinstall the hook. Users upgrading from older graphify versions no longer get stuck with a broken hook format that causes PreToolUse errors (#182)
• Windows CRLF: rationale node labels no longer contain bare carriage returns when processing files with CRLF line endings. This was silently breaking Obsidian exports on Windows and WSL (#176)
• Windows skill: skill-windows.md now includes --wiki, --obsidian-dir, and --directed which were missing compared to the main skill (#177)

Upgrade with pip install --upgrade graphifyy

Fixed graphify install --platform gemini not copying the skill file to ~/.gemini/skills/graphify/SKILL.md. The /graphify trigger was not working without it. Upgrade with pip install --upgrade graphifyy

Security fix for the MCP server path validation introduced in 0.3.25. The previous fix for #170 used the graph file's own parent directory as the validation base, which meant any path would pass validation against itself. The MCP server now checks that the path exists and ends in .json instead. Upgrade with pip install --upgrade graphifyy

What's new

Fixes

• graphify install --platform gemini and --platform cursor no longer error — both now route to their correct install functions (#171)
• MCP server (serve.py) now works when graph is outside cwd — validate_graph_path receives the correct base (#170)
• MCP call_tool() handler catches exceptions and returns graceful error strings instead of crashing the stdio loop (#163)
• .graphifyignore discovery now walks parent directories up to the .git boundary, matching .gitignore behavior — subdirectory scans inherit root patterns (#168)

New platforms

• Aider — graphify install --platform aider + graphify aider install/uninstall (#74)
• GitHub Copilot CLI — graphify install --platform copilot + graphify copilot install/uninstall (#134)

New features

• --directed flag — builds a DiGraph preserving edge direction (source→target); cluster() converts to undirected internally for Leiden (#125)
• Frontmatter-aware cache for Markdown — .md files hash only the body, so metadata-only changes (reviewed, status, tags) no longer invalidate the cache (#131)

What's new

• Fix: codex/opencode install idempotency — graphify codex install no longer exits early when AGENTS.md already has the graphify section. Partial installs with a missing .codex/hooks.json now recover on re-run (#153)

Install / upgrade

pip install --upgrade graphifyy

What's new

• Add: Gemini CLI support — graphify gemini install writes a GEMINI.md section and a BeforeTool hook in .gemini/settings.json that fires before file-read tool calls (#105)
• Add: sponsor nudge at pipeline completion — all skill files now print a one-line sponsor link after a fresh build (not on --update runs)

Install / upgrade

pip install --upgrade graphifyy
graphify gemini install   # Gemini CLI users

Full changelog: CHANGELOG.md

What's new

• Add: Cursor support — graphify cursor install writes .cursor/rules/graphify.mdc with alwaysApply: true; graphify cursor uninstall removes it (#137)
• Fix: _rebuild_code() KeyError — git hooks were silently failing on every commit because detected[FileType.CODE] didn't match detect()'s actual return shape (#148)
• Fix: crash on NetworkX 3.2.x — node_link_data() now falls back gracefully on older NetworkX, same shim already used everywhere else (#149)
• Fix: official package clarification in README — graphifyy is the only official PyPI package (#129)

Install / upgrade

pip install --upgrade graphifyy

Re-run graphify hook install to get the fixed git hooks.

Full changelog: CHANGELOG.md

What's new

• Fix: Codex hook JSON schema — systemMessage now placed at the top level of the hook output, not inside hookSpecificOutput. codex-cli 0.118.0+ uses additionalProperties: false on the inner object so the old placement caused a validation error (#138)
• Fix: git hooks now use #!/bin/sh instead of #!/bin/bash — Git for Windows ships sh.exe, not bash, so hooks were silently skipped after every commit on Windows (#140)

Install / upgrade

pip install --upgrade graphifyy

Then re-run graphify codex install to get the fixed hook, and graphify hook install to get the fixed git hook.

Full changelog: CHANGELOG.md

What's new

• Fix: XSS in interactive HTML graph — node labels, file types, community names, source files, and edge relations now HTML-escaped before innerHTML injection; neighbor link onclick uses JSON.stringify instead of raw string interpolation
• Add: OpenCode tool.execute.before plugin — graphify opencode install now writes .opencode/plugins/graphify.js and registers it in opencode.json, firing the graph reminder before bash calls (equivalent to Claude Code's PreToolUse hook) (#71)
• Fix: AST call edges now EXTRACTED/1.0 instead of INFERRED/0.8 — tree-sitter call resolution is deterministic, not probabilistic (#127)
• Fix: tree-sitter>=0.23.0 pinned and _check_tree_sitter_version() guard added — stale environments now get a clear error with upgrade instructions instead of a cryptic TypeError (#89)

Install / upgrade

pip install --upgrade graphifyy

Full changelog: CHANGELOG.md

Fix

• Install step now tries a plain pip install first and only falls back to --break-system-packages if that fails. Homebrew and PEP 668 managed environments will use the clean path and never see the flag in most cases (#126)

Install

pip install graphifyy==0.3.19 && graphify install

Fixes

• Watch + .graphifyignore: --watch mode now respects .graphifyignore — _rebuild_code was calling collect_files() directly instead of detect(), bypassing ignore patterns (#120)
• Codex PreToolUse hook: now uses systemMessage instead of additionalContext — Codex does not support additionalContext and was returning an error (#121)
• Trae link: corrected from trae.com to trae.ai across all READMEs (#122)
• HTML graph click detection: hover-tracking (hoveredNodeId) replaces unreliable vis.js click params — small/dense nodes now reliably selectable (#82)
• .graphify_python persistence: mkdir -p graphify-out runs before writing .graphify_python; file no longer deleted in Step 9 cleanup — pipx users no longer hit ModuleNotFoundError on follow-up commands (#93)
• Trae skill packaging: skill-trae.md added to pyproject.toml package-data — Trae users were getting ModuleNotFoundError after pip install (#102)
• Extension drift: analyze.py and watch.py now import extension sets from detect.py instead of local copies — Swift, Lua, Zig, PowerShell, Elixir, JSX, Julia, Objective-C files were being misclassified as documents (#109)

Additions

• graphify save-result CLI subcommand — saves Q&A results to memory dir; replaces inline Python blocks in all 6 skill files (#114)
• Korean README added (README.ko-KR.md) (#112)
• Interpreter guard section in skill.md for when graphify-out/ is deleted before follow-up commands (#93)

Refactors

• Dead build_graph() function removed from cluster.py (#109)

Install

pip install graphifyy==0.3.18 && graphify install

• Add: Julia (.jl) support — modules, structs, abstract types, functions, short functions, using/import, call edges, inherits edges (#98)
• Fix: Semantic extraction chunks now group files by directory — reduces missed cross-chunk relationships (#65)
• Fix: `tree-sitter>=0.21` now pinned — prevents silent empty AST output on older tree-sitter versions (#52)
• Add: Progress output every 100 files during AST extraction on large projects (#52)

• Fix: `graphify query`, `serve`, and `benchmark` now work on NetworkX < 3.4 — version-safe shim for `node_link_graph()` at all call sites (#95)
• Fix: `.jsx` files now detected and extracted via the JS extractor (#94)
• Fix: `.graphify_python` no longer deleted in Step 9 cleanup across all 6 skill files — pipx users no longer hit `ModuleNotFoundError` on follow-up commands (#92)

0.3.15

• Feat: Trae and Trae CN platform support (graphify install --platform trae / trae-cn)
• Fix: skill-droid.md was missing from PyPI package data — Factory Droid users couldn't install the skill
• Fix: XSS in HTML legend — community labels now HTML-escaped before innerHTML injection
• Fix: Shebang allowlist validation in hooks.py and all skill files — prevents metacharacter injection
• Fix: louvain_communities() kwargs now inspected at runtime for cross-version NetworkX compatibility
• Fix: pipx installs now detected correctly in git hooks (reads shebang from graphify binary)
• Fix: graspologic ANSI escape codes no longer corrupt PowerShell 5.1 scroll buffer
• Docs: Japanese README added
• Docs: graph.json + LLM workflow example added to README

0.3.14

• Fix: graphify codex install now also writes a PreToolUse hook to .codex/hooks.json so the graph reminder fires before every Bash tool call (#86)
• Fix: --update now prunes ghost nodes from deleted files before merging new extraction (#51)

0.3.13

• Fix: PreToolUse hook now outputs additionalContext JSON so Claude actually sees the graph reminder before Glob/Grep calls (#83)
• Fix: Go AST method receivers and type declarations now use package directory scope, eliminating disconnected duplicate type nodes across files in the same package (#85)
• Fix: PDFs inside Xcode asset catalogs (.imageset, .xcassets) are no longer misclassified as academic papers (#52)
• Fix: _resolve_cross_file_imports is now guarded with if py_paths and wrapped in try/except so a Python parser crash can't abort extraction for non-Python files (#52)
• Fix: Skill intermediate files (.graphify_*.json) now live in graphify-out/ instead of project root, preventing git pollution (#81)

0.3.12

• Fix: sanitize_label was double-encoding HTML entities in the interactive graph (< instead of <) — removed html.escape() from sanitize_label; callers that inject directly into HTML now call html.escape() themselves (#66)
• Fix: --wiki flag missing from skill.md usage table (#55)

What's new

Fix: Louvain fallback hang on large sparse graphs (#48)

• Added max_level=10, threshold=1e-4 to NetworkX Louvain fallback
• Prevents indefinite hang on large sparse graphs (49K+ nodes, low density)
• Equivalent community quality on typical corpora
• Only affects users without graspologic installed (Leiden is unaffected)

What's new

Fix: Windows Unicode crash (#47)

• Replaced arrow character with -> in all print statements — fixes UnicodeEncodeError on cp1252/Windows consoles

Add: Skill staleness warning (#46)

• graphify install now writes a version stamp alongside the skill file
• Every graphify command checks all platform skill locations and warns if the installed skill is from an older version
• Works across Claude Code, Windows, Codex, OpenCode, OpenClaw, Factory Droid

What's new

Symlink support (#33)

• detect() and collect_files() now accept follow_symlinks=True to traverse symlinked files and directories
• Circular symlink cycle detection prevents infinite recursion
• watch.py now uses collect_files() for consistency

Docs

• Codex skill trigger is $graphify . not /graphify . (#36)

Tests

• 367 tests passing (up from 362)

What's new

C# inheritance extraction (#45)

• DataProcessor : IProcessor now emits inherits edges
• Handles both simple bases (identifier) and generic bases (IList<T> → generic_name)
• Zero false positives — tested on 6,100-file Unity/VaM codebase producing 3,957 edges

graphify query CLI command (#43)

• Query the knowledge graph without Claude Code:

  graphify query "what connects attention to the optimizer?"
  graphify query "what is CfgNode?" --dfs
  graphify query "show auth flow" --budget 500 --graph path/to/graph.json
  

• BFS by default, --dfs for path tracing, --budget N to cap output tokens

Tests

• 362 tests passing (up from 360)

What's new

Objective-C support (#34)

• .m and .mm files now extracted via tree-sitter
• @interface, @implementation, @protocol → class/protocol nodes
• Method declarations and definitions → method nodes with method edges
• #import directives → imports edges
• Superclass and protocol conformance → inherits edges
• Message expressions → calls edges (INFERRED)

--obsidian-dir flag (#37)

• Write Obsidian vault to a custom path: /graphify ./raw --obsidian --obsidian-dir ~/vaults/myproject

Bug fixes

• Semantic cache: relative paths from subagents now resolved against corpus root — was causing only ~25% of files to cache correctly
• Validation: file_type: "rationale" no longer triggers 75 warnings per run

Tests

• 360 tests passing (up from 354)

What's new

• Elixir language support — .ex and .exs files now go through AST extraction. Extracts defmodule, def/defp, alias/import/require/use, and the call graph between functions. Same EXTRACTED/INFERRED tagging as every other language.

What's new

• .docx and .xlsx support — Office documents now work in graphify. Install the optional extra:

  pip install graphifyy[office]
  

  • .docx: converted to markdown via python-docx (headings, lists, tables preserved)
  • .xlsx: each sheet converted to a markdown table via openpyxl
  • Converted files are saved to graphify-out/converted/ as .md sidecars — Claude reads them as normal documents
  • Graceful fallback if [office] is not installed: file is skipped with an install hint

What's new

• Fix UnicodeDecodeError on Windows GBK locale — all read_text() and write_text() calls in graphify/__main__.py now explicitly pass encoding="utf-8". Windows systems with non-UTF-8 default locales (e.g. GBK in Chinese Windows) no longer crash on graphify install.

What's new

• Windows support — graphify install now auto-detects Windows and installs a PowerShell-compatible skill variant (skill-windows.md). All bash-specific commands replaced: which/head/tr → Python detection, rm -f → Remove-Item, mkdir -p → New-Item, python3 → python
• Manual override: graphify install --platform windows on any platform

Zig and PowerShell are now supported. Functions, structs, enums, unions, and imports for Zig. Functions, classes, methods, and using statements for PowerShell.

Factory Droid is now a supported platform. graphify install --platform droid and graphify droid install work the same as the other platforms.

.graphifyignore — add a file to exclude folders and files from the graph. Same syntax as .gitignore. Useful for keeping node_modules, vendor, dist, and generated files out of the graph.

What is next

v0.4.0 brings local embeddings via quantized Gemma 4, no API cost for semantic similarity. Also planned: more /graphify add sources, graph diff across runs, and worked examples for Codex and OpenCode.

Fourteen bugs fixed across two community code reviews.

Git hooks now exit with a non-zero code when a rebuild fails. Previously the hook caught the exception and exited 0, so git assumed everything was fine and your graph would go stale with no warning.

The /graphify query token budget was overshooting by about 3x. The character-to-token ratio has been corrected.

Deleted files are now tracked in incremental mode. If you remove a file and run /graphify --update, its nodes are flagged for cleanup instead of staying in the graph as ghost nodes.

Cache writes are now atomic. A crash mid-write no longer leaves a corrupt cache entry that silently causes a full re-extraction on the next run.

/graphify add tweet fetching was bypassing SSRF protections and making raw HTTP requests. It now routes through the same safe fetch layer as everything else.

Neo4j Cypher export was vulnerable to injection via backslash sequences in node labels. Escaping is now correct.

The HTML visualization was embedding AI-generated community labels directly into innerHTML without sanitization. Labels are now sanitized before rendering.

Graph diff was reporting phantom additions and removals on undirected graphs because edge key ordering was non-deterministic. Edge keys are now normalized.

The file cache was keyed only on content hash, so two files with identical content would share a cache entry with wrong source metadata baked in. The file path is now included in the key.

Hook uninstall used a naive string split that could corrupt other hooks sharing the same file. Uninstall now uses paired start and end markers.

What is next

v0.4.0 brings local embeddings via quantized Gemma 4, no API cost for semantic similarity. Also planned: more /graphify add sources, graph diff across runs, and worked examples for Codex and OpenCode.

What's new in v0.3.0

Multi-platform support

graphify now works on Claude Code, Codex, OpenCode, and OpenClaw.

Platform	Skill install	Always-on
Claude Code	graphify install	graphify claude install
Codex	graphify install --platform codex	graphify codex install
OpenCode	graphify install --platform opencode	graphify opencode install
OpenClaw	graphify install --platform claw	graphify claw install

Each platform gets its own optimized skill file. Codex uses spawn_agent, OpenCode uses @mention, OpenClaw uses sequential extraction.

Always-on project hooks for all platforms

• graphify claude install adds CLAUDE.md rules and a PreToolUse hook so Claude checks the graph before every file search (Claude Code only)
• graphify codex install / opencode install / claw install writes the same rules to AGENTS.md for always-on graph-first behaviour on those platforms

MIT license

graphify is now open source under the MIT license.

Bug fixes

• Hyperedges were silently dropped when merging multiple extraction passes, now fixed

---

What's next (v0.4.0)

• Local embeddings via quantized Gemma 4 (Q4/Q8, runs locally, no API cost) to generate semantic similarity edges across all nodes without any API calls. Opt-in via pip install graphifyy[embeddings]
• More sources for /graphify add including YouTube transcripts, GitHub issues, and Notion pages
• Graph diff showing what changed between two graph snapshots (new nodes, dropped edges, community shifts)
• Worked examples for Codex, OpenCode, and OpenClaw with honest evaluations

---

Install

pip install graphifyy==0.3.0
graphify install                          # Claude Code
graphify install --platform codex         # Codex
graphify install --platform opencode      # OpenCode
graphify install --platform claw          # OpenClaw

• Add: graphify claude install — writes graphify section to local CLAUDE.md + PreToolUse hook in .claude/settings.json
• Add: graphify claude uninstall — removes section and hook
• Add: graphify hook install — installs post-commit and post-checkout git hooks (platform-agnostic)
• Add: graphify hook uninstall / hook status
• Add: graphify benchmark CLI command
• Fix: node deduplication documented at all three layers

What's new in v0.2.0

Mine the "why"

Code comments and docstrings are now first-class graph nodes. Module, class, and function docstrings are extracted automatically. Inline comments tagged # NOTE:, # IMPORTANT:, # HACK:, # WHY:, # RATIONALE:, # TODO:, # FIXME: become rationale_for nodes connected to the code entity they explain. Doc files now extract design rationale and architectural decisions too - not just structure, but why things were built that way.

Confidence scores on every edge

Every INFERRED edge now carries a confidence_score (0.0-1.0). EXTRACTED edges are always 1.0. INFERRED edges score 0.4-0.9 based on how certain the model was. AMBIGUOUS edges score 0.1-0.3. You know not just what was guessed but how confident.

Semantic similarity edges

Cross-file conceptual links with no structural connection. Two functions solving the same problem without calling each other, a class in code and a concept in a paper describing the same algorithm. Tagged semantically_similar_to with a confidence score.

Hyperedges

Group relationships connecting 3+ nodes that pairwise edges cannot express. All classes implementing a shared protocol, all functions in an auth flow, all concepts from a paper section forming one coherent idea. Stored in a top-level hyperedges array in graph.json.

Git hooks - post-checkout added

graphify hook install now installs both post-commit (existing) and post-checkout (new). The graph rebuilds automatically when you switch branches, not just when you commit.

Claude Code PreToolUse hook

graphify claude install now also writes a .claude/settings.json hook that fires before every Glob and Grep call. If a graphify knowledge graph exists, Claude is reminded to check it before searching raw files. Ensures the graph is actually used during research, not bypassed.

Obsidian output opt-in

Obsidian vault generation (--obsidian) is now opt-in. Default output is graph.html + graph.json + GRAPH_REPORT.md. Large repos were generating thousands of files by default.

Upgrade

pip install --upgrade graphifyy
graphify install