19 Commits

Author SHA1 Message Date
Safi 23f598f3a0 fix MultiGraph crash, hollow LLM response retry, and skill --help (#796, #795, #792)
#796: add edge_data()/edge_datas() helpers in build.py that tolerate
MultiGraph/MultiDiGraph; replace all G.edges[u,v] 2-tuple call sites in
__main__.py, serve.py, wiki.py, export.py, analyze.py, benchmark.py;
fix same pattern in 10 skill file inline heredocs

#795: all 12 skill files now short-circuit on /graphify --help or -h
and print the Usage block without running any pipeline steps

#792 (hollow response): add _response_is_hollow() predicate in llm.py;
when Ollama (or any backend) returns empty/null/whitespace content or a
parsed result with no nodes/edges, rewrite finish_reason="length" so
_extract_with_adaptive_retry bisects the chunk instead of silently
dropping it; applied to _call_openai_compat, _call_claude, _call_bedrock

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-05-09 21:17:25 +01:00
Arjun Pramanik a15cb36dfc fix(skills): rewrite YAML descriptions from pipeline-only to trigger-oriented (#774)
10 skill-*.md files had descriptions that only described what graphify does (input->pipeline->output), not when agents should use it. This meant skills never loaded proactively on codebase questions.

Changed to hybrid descriptions that retain the pipeline summary but add trigger conditions: 'Use when user asks any question about a codebase, project content, architecture, or file relationships'.
2026-05-07 20:07:49 +01:00
Safi d819827ea2 fix #655 #656 #657 #658: cache dir crash, sanitize_label None, rationale file_type, token counting
#655 and #656 already fixed in cache.py and security.py.
#657: add rationale to file_type schema in all 12 skill variants; warn against inventing concept.
#658: add explicit chunk-merge step with token summation before save_semantic_cache.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-05-02 17:10:20 +01:00
Safi 4563b043f2 Fix 6 bugs: ID collisions, path portability, alias resolution, HTML controls, desync guard, rationale prompt
- #550: _file_stem() includes parent dir to prevent node ID collisions for same-named files
- #555: extract() relativizes source_file paths before returning for cross-machine portability
- #562: to_json() returns bool; _rebuild_code() writes report/html only if json succeeded
- #563: skill prompts store rationale as node attribute, not separate node; enforce calls direction
- #566: Show All / Hide All buttons added to HTML community panel
- #575: _import_js() resolves tsconfig.json compilerOptions.paths aliases before external fallback

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-27 21:44:22 +01:00
Safi 64f38acf3e implement #488 #482 #472 #490: legacy schema canonicalization, Java inheritance, aggregated HTML viz, check-update subcommand 2026-04-22 23:28:09 +01:00
Safi 52ad45bd27 fix #504: export PYTHONUTF8=1 in Step 1 to prevent garbled CJK output on Windows 2026-04-22 09:28:26 +01:00
Safi 86d6d9317f fix #498: opencode config to .opencode/opencode.json, fix skill temp file paths, add --wiki step 2026-04-22 09:19:53 +01:00
Safi 9c3d0fa517 Fix chunk temp files not cleaned up after graph build
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-21 22:21:37 +01:00
Safi 1dd4c41413 Remove Claude-specific language from platform skill files (0.4.8)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-12 21:18:20 +01:00
Safi cede5b6ca6 Fix #188: collect_files() now respects .graphifyignore; fix #195: skill.md requires general-purpose subagent type for extraction dispatch
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-10 18:41:48 +01:00
Safi 2a7d2f4fdc Remove Anthropic API call from transcribe.py - agent generates Whisper prompt itself
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-10 15:55:22 +01:00
Safi cf5d721bee Add video/audio corpus support with yt-dlp download and Whisper transcription
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-10 15:40:18 +01:00
Safi df26f85648 Add Gemini CLI support and sponsor nudge at pipeline completion (#105)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-09 17:40:45 +01:00
Safi be2058d5d5 Fix install step: try plain pip before --break-system-packages (#126)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-09 08:40:17 +01:00
Safi 91cbc74bb1 release 0.3.18: fix watch .graphifyignore, codex hook, trae link, Korean README, save-result CLI 2026-04-09 08:09:36 +01:00
Safi 7ff7bd2e8f v0.3.17: Julia support, smarter chunking, tree-sitter pin, progress output
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-08 22:54:18 +01:00
Safi 1896379283 v0.3.16: NetworkX <3.4 compat, .jsx support, pipx follow-up fix
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-08 22:27:08 +01:00
Safi 359ef86dd4 fix: shebang allowlist validation in all skill files 2026-04-08 19:57:33 +01:00
Safi 38d967fc63 feat: multi-platform skill files and install routing (Codex, OpenCode, OpenClaw) 2026-04-06 16:06:31 +01:00