Plan, search, extract, and assemble research reports from disk.
npx skills add https://github.com/arozumenko/sdlc-skills --skill deep-research-arozumenko
Please help me install this Agent Skill. Skill: deep-research Source: https://github.com/arozumenko/sdlc-skills/tree/main/skills/deep-research Command: npx skills add https://github.com/arozumenko/sdlc-skills --skill deep-research-arozumenko
It eliminates unreliable, memory-only research by running a repeatable workflow that writes plans, findings, and checkpoints to disk so you can resume and assemble final reports without losing work.
Ask the AI to fact-check this document using deep-research and save a disk checkpointed workspace, then produce a detailed verdict report with cited evidence.
High-intent search queries and answers about installing and using this skill.
Fact-checking with evidence requires extracting claims, verifying each against sources, and saving disk checkpoints. This workflow writes plans, findings, and verdicts to disk, ensuring you can resume and assemble final reports without losing work.
Trend analysis needs a repeatable workflow that identifies players, adoption signals, and emerging patterns using sources and publication dating. A disk-first research approach writes findings to disk, eliminating unreliable memory-only research by maintaining checkpointed progress.
A disk-first research workflow plans, searches, extracts, and assembles results from disk to produce deep topic decomposition. It creates balanced analysis with tradeoffs and clearly separated recommendations by checkpointing each phase to preserve intermediate findings.
Yes, automated research workflows require Tavily tools like tavily_search and tavily_extract for source retrieval and content extraction. Context7 tools are optionally supported for technical grounding to resolve library IDs and query documentation.
Research workflows support both SWOT-style analysis and fact verification across three modes. Trend research identifies market changes, topic analysis produces balanced decomposition with tradeoffs, and fact-checking extracts claims to verify with evidence and impartial verdicts.
Checkpointing matters for evidence-based research because it writes plans, findings, and intermediate results to disk during the research process. This prevents data loss inherent in memory-only research, allowing you to resume work and assemble final reports reliably.
auxi-wardrobeAutomate documentation discovery using llms.txt, context7 API, and GitHub.
wangbinquanDetect mismatches between design documents and code across nine consistency dimensions.
TheTesseractAISearch the web and fetch page content as markdown.
XSpoonAiAutomate topic research and synthesize findings into a structured report.
