Repository Issue Activity (beta)

#462 Security: Path traversal in fetch_config config_name (CWE-22) - arbitrary directory deletion, repository destruction, attacker-controlled clone

#453 Unified multi-source build: PDF content silently dropped from final output, SKILL.md becomes AI-hallucinated filler with no grounded content

#452 Feature Suggestion: Optional token metering & paid API key support via `neuforge-pay`

#434 [BUG] PDF image extraction fails for vector figures — saves fill-pattern tiles instead of diagrams

#228 feat(quality): Add readability scoring for skills (G2.3)

#96 [F1.4] Add HTML parser fallback (lxml → html5lib)

#393 [FEATURE] SQLite index layer over skill markdown (script-queryable, token-efficient)

#97 [F1.5] Add network retry with exponential backoff

#169 fix: Reduce token bloat - Stop scraping closed issues and unnecessary metadata

#229 enhance(quality): Refine completeness check patterns to reduce false positives

#260 🌏 帮助改进中文 README 翻译 / Help Improve Chinese README Translation

#425 FactoryDetector matches 'get' as a substring: every getter inflates Factory confidence (plain POJO -> Factory 0.90); Singleton unreachable in Java; dependency graph emits 0 edges

#4 Add an example project to see how a successfull project looks like.

#92 [E2.6] Add retry logic for network failures in MCP

#41 [B2.7] Add MCP tool scrape_docx

#167 feat: Version tracking for official vendor tutorials - Auto-detect API changes

#172 [FEATURE] How about one online market?

#205 [A1.8] Add smart skill detection and auto-install

#153 🔥🔥🔥中文视频教程

#138 [E1.5] Add scrape_xlsx MCP tool

#137 [E1.4] Add scrape_docx MCP tool

#171 But Claude Sonnet 4.5 and ChatGPT seem to disagree about COSTS (Shared Chats)

#103 [F2.5] Preserve local annotations/edits

#111 [G2.3] Add readability scoring

#106 [G1.3] Create auto_detect_selectors.py (AI-powered)