[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-84005":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":17,"stars30d":17,"stars90d":16,"forks30d":16,"starsTrendScore":17,"compositeScore":18,"rankGlobal":10,"rankLanguage":10,"license":19,"archived":20,"fork":20,"defaultBranch":21,"hasWiki":22,"hasPages":22,"topics":23,"createdAt":10,"pushedAt":10,"updatedAt":34,"readmeContent":35,"aiSummary":10,"trendingCount":16,"starSnapshotCount":16,"syncStatus":36,"lastSyncTime":37,"discoverSource":38},84005,"sisyphus-academica","argahv\u002Fsisyphus-academica","argahv","Open-source research pipeline — literature review, novelty generation, citation verification, and adversarial review.","https:\u002F\u002Fargahv.github.io\u002Fsisyphus-academica\u002F",null,"Python",56,12,52,14,0,4,3.34,"MIT License",false,"main",true,[24,25,26,27,28,29,30,31,32,33],"academic-writing","arxiv","citation-verification","latex","literature-review","machine-learning","multi-agent","novelty","open-source","research","2026-06-12 02:04:37","\u003Cdiv align=\"center\">\n\n# Sisyphus Academica — The Research Paper Writing Army\n\n**20+ specialized agents. 6 novelty engines. 10 adversarial reviewers. Zero hallucinated citations. Zero AI-isms.**\n\n⭐ **If you write research papers, star this repo — it will save you weeks of work.**\n\n[![CI](https:\u002F\u002Fgithub.com\u002Fargahv\u002Fsisyphus-academica\u002Factions\u002Fworkflows\u002Fci.yml\u002Fbadge.svg)](https:\u002F\u002Fgithub.com\u002Fargahv\u002Fsisyphus-academica\u002Factions\u002Fworkflows\u002Fci.yml)\n[![License: MIT](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-MIT-yellow.svg)](LICENSE)\n[![Python 3.10+](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fpython-3.10+-blue.svg)](pyproject.toml)\n[![GitHub Stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fargahv\u002Fsisyphus-academica?style=social)](https:\u002F\u002Fgithub.com\u002Fargahv\u002Fsisyphus-academica)\n[![Star History](https:\u002F\u002Fapi.star-history.com\u002Fsvg?repos=argahv\u002Fsisyphus-academica&type=Date)](https:\u002F\u002Fstar-history.com\u002F#argahv\u002Fsisyphus-academica)\n\n\u003C\u002Fdiv>\n\nNot a writing assistant. Not a chatbot with a LaTeX plugin. A **self-coordinating swarm** of 20+ specialized agents that produces publication-ready research papers with **genuine novelty, zero hallucinated citations, and no detectable AI-written patterns.**\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fargahv\u002Fsisyphus-academica.git && cd sisyphus-academica\nbash install.sh\n# Select \"research-director\" → \"write a paper about [topic]\"\n```\n\n---\n\n## CLI Tools (No Agent Required)\n\nThe Python CLI works standalone — no OpenCode or agent platform needed:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fargahv\u002Fsisyphus-academica.git && cd sisyphus-academica\npip install -e .\n\nsisyphus demo              # Interactive pipeline demo (no API keys)\nsisyphus search QUERY      # Search 4 academic APIs in parallel\nsisyphus verify FILE       # Verify citations in a paper JSON\nsisyphus bibtex DOI        # Generate BibTeX from a DOI\nsisyphus configure         # Set up API keys interactively\n```\n\n---\n\n## Portable Agent Skills (Works with Any Agent)\n\nThe novelty engines and reviewer personas are packaged as **standalone agent skills** — drop them into any agent that reads SKILL.md (Claude Code, Codex, Cursor, Gemini CLI, OpenCode, and more):\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fargahv\u002Fsisyphus-academica.git\n# Manual: copy individual skill directories\ncp -r skills\u002Fcontrarian ~\u002F.claude\u002Fskills\u002F\n# Or use npx (see below)\n```\n\nOr install any skill with a single command:\n\n```bash\nnpx skills add argahv\u002Fsisyphus-academica          # Browse and select interactively\nnpx skills add argahv\u002Fsisyphus-academica -s '*'   # Install all 17 skills\nnpx skills add argahv\u002Fsisyphus-academica -s contrarian  # Install one skill\n```\n\nThen invoke directly in your agent:\n\n```\n\u002Fcontrarian \"The claim: 'Attention is all you need'\"\n\u002Fcross-pollinator \"Problem: How to reduce LLM hallucination\"\n\u002Fheretic \"Paper: 'Scaling Laws for Neural Language Models'\"\n```\n\n**6 novelty engines + 10 reviewers + academic humanizer = 17 portable skills**.\n\n> 💡 **Prefer a standalone repo?** The 6 novelty engines and 5 reviewer personas are also available as a dedicated skill pack at **[argahv\u002Fnovelty-skills](https:\u002F\u002Fgithub.com\u002Fargahv\u002Fnovelty-skills)** — install via `npx skills add argahv\u002Fnovelty-skills`.\n\n---\n\n## For Humans\n\n**Strongly recommended: let an LLM agent install this for you.** The setup involves API key configuration, agent deployment across 25 specialized agents, voice profile calibration, and provider selection — humans fat-finger these. An LLM agent reads the full guide and walks every step correctly.\n\nThe agent will greet you with `oMoMoMoMo...` and ask **one question at a time**:\n1. Which platform? (OpenCode recommended)\n2. Do you have oh-my-openagent?\n3. Claude subscription? (Pro \u002F Max 20x \u002F None)\n4. OpenAI subscription? (Plus \u002F None)\n5. Semantic Scholar API key? (free — boosts rate limits)\n6. LaTeX for PDF? (or Docker \u002F skip for .tex only)\n7. Writing sample for voice calibration? (optional)\n\nThen it installs, configures, and verifies everything automatically.\n\nPaste this prompt into Claude Code, AmpCode, Cursor, or any agent:\n\n```\nInstall and configure Sisyphus Academica by following the instructions here:\nhttps:\u002F\u002Fraw.githubusercontent.com\u002Fargahv\u002Fsisyphus-academica\u002Fmain\u002Fdocs\u002Fguide\u002Finstallation.md\n```\n\n### Manual install — one line, no agent needed\n\nIf you prefer to run the installer yourself:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fargahv\u002Fsisyphus-academica.git && cd sisyphus-academica\nbash install.sh\n```\n\nThe installer will ask you the same 6 questions and configure everything automatically. Run with `--yes` to skip all prompts and use defaults:\n\n```bash\nbash install.sh --yes\n```\n\n---\n\n## The Numbers (Not Claims)\n\n| Metric | GPT-4 \u002F NotebookLM | Sisyphus Academica |\n|---|---|---|\n| Papers surveyed per run | 10-50 | **500+** (5 parallel scouts) |\n| Citation hallucination rate | ~30-40% | **0%** (verified against 2+ APIs) |\n| Novelty generation | None (\"what's the gap?\") | **6 engines × 50+ hypotheses** |\n| Adversarial review | None | **10 distinct personas** |\n| AI-text detection | Post-hoc (chatgpt.com paste) | **Generation-time (41 patterns)** |\n| Voice calibration | None | **Learns from your writing samples** |\n| Output format | Raw text \u002F Google Doc | **LaTeX PDF with verified BibTeX** |\n| Time to first draft | 5-20 min | **30 min - 4 hours** (reviewed) |\n\n**Pipeline stats from our SIREN paper** ([view output](examples\u002Fsiren-paper\u002F)):\n100+ papers surveyed → 6 novelty engines → 50+ hypotheses → 10 adversarial reviewers → 4 revision rounds → **13-page PDF, 26 verified citations, 3 publication-ready figures, 0 AI-pattern violations, 0 em dashes, 0 hallucinated references.**\n\n---\n\n## Quick Start\n\n```bash\n# 1. Clone + install\ngit clone https:\u002F\u002Fgithub.com\u002Fargahv\u002Fsisyphus-academica.git\ncd sisyphus-academica\nbash install.sh\n\n# 2. Configure API keys (free)\ncp .env.example .env\n# Add your Semantic Scholar API key + CrossRef email\n\n# 3. Write a paper\n# OpenCode → agent tab → select \"research-director\"\n# → type: \"write a paper about transformer efficiency\"\n```\n\n**Prerequisites:** OpenCode (or compatible agent platform), Python 3.10+, LaTeX (optional, for PDF).\n\n**Provider-agnostic** — works with any OpenAI-compatible or Anthropic API. Edit `config\u002Fagent-config.json` to switch models:\n\n```json\n{\n  \"agents\": {\n    \"research-director\": {\n      \"model\": \"anthropic\u002Fclaude-opus-4\",\n      \"fallback_models\": [{\"model\": \"anthropic\u002Fclaude-sonnet-4\"}]\n    }\n  }\n}\n```\n\n---\n\n## Architecture\n\n```\n                     ┌──────────────────────────────┐\n                     │  Research Director            │\n                     │  (orchestrator)               │\n                     └────────┬─────────────────────┘\n                              │\n         ┌────────────────────┼─────────────────────────┐\n         ▼                    ▼                          ▼\n   ┌───────────┐    ┌────────────────┐    ┌──────────────────────┐\n   │ Literature│    │ 6 Novelty      │    │ Gap Analyzer         │\n   │ Scout ×5  │    │ Engines        │    │ + Methodology        │\n   └───────────┘    └────────────────┘    └──────────────────────┘\n                              │\n         ┌────────────────────┼─────────────────────────┐\n         ▼                    ▼                          ▼\n   ┌─────────────────────────────────────────────────────────────┐\n   │  Parallel Writing Swarm (5 agents + 41 Humanizer patterns)   │\n   └──────────────────────────┬──────────────────────────────────┘\n                              │\n   ┌─────────────────────────────────────────────────────────────┐\n   │  Verifier: Citations × Statistics × AI-Pattern Detection    │\n   └──────────────────────────┬──────────────────────────────────┘\n                              │\n   ┌─────────────────────────────────────────────────────────────┐\n   │  Adversarial Review: 10 reviewer personas (parallel)        │\n   └──────────────────────────┬──────────────────────────────────┘\n                              │\n   ┌─────────────────────────────────────────────────────────────┐\n   │  Style Auditor: Humanizer certification, em dash zero check │\n   └──────────────────────────┬──────────────────────────────────┘\n                              │\n   ┌─────────────────────────────────────────────────────────────┐\n   │  Formatter: LaTeX template, BibTeX, figures, PDF            │\n   └─────────────────────────────────────────────────────────────┘\n```\n\n---\n\n## The Novelty Engines (The Moat)\n\nSix engines that think like no human can:\n\n1. **The Contrarian** — Inverts every well-established claim in the field. Generates 10 counter-hypotheses.\n2. **The Cross-Pollinator** — Imports solutions from 15 distant fields (astrodynamics, epidemiology, music theory, immunology, 15th-century shipbuilding...).\n3. **The Assumption Excavator** — Finds unstated assumptions and tests what breaks if they're false.\n4. **The Counterfactual Generator** — Rewrites the field's history without the most-cited papers.\n5. **The Paradox Sifter** — Cross-references every \"Limitations\" section to find ignored contradictions.\n6. **The Heretic** — **Crown jewel.** Generates 50 wild hypotheses from title+abstract alone, scores each against the actual paper, and finds the \"haunting idea\" — what the paper *should have been*.\n\n---\n\n## The Adversarial Reviewers\n\nEach paper is independently reviewed by 10 distinct personas running in parallel. All 10 must pass before formatting.\n\n| Persona | Focus | Typical Critique |\n|---------|-------|-----------------|\n| **Theorist** | Formal proofs, mathematical rigor | \"Where's the formal proof?\" |\n| **Empiricist** | Experimental design, baselines | \"Your baseline is wrong\" |\n| **Pragmatist** | Practical applicability | \"Does this matter in practice?\" |\n| **Skeptic** | Default: results are wrong | \"Show me error bars\" |\n| **Historian** | Prior art, citation accuracy | \"This was done in 1972\" |\n| **Methodologist** | Statistical methodology | \"Your test assumes normality\" |\n| **Ethicist** | Societal implications | \"What are the downsides?\" |\n| **Competitor** | Novelty relative to existing work | \"Minor mod of our 2023 paper\" |\n| **Student** | Clarity and accessibility | \"I don't understand section 3\" |\n| **Dreamer** | \"What if you went further?\" | \"You stopped too early\" |\n\n---\n\n## Quality Gates\n\nEvery paper passes through 5 hard gates. **If any gate fails, the paper goes back to revision.**\n\n1. **Citation Verification** — Every reference checked against Semantic Scholar + CrossRef APIs. Must be found in 2+ sources. No exceptions.\n2. **Statistical Audit** — Every p-value, effect size, sample size, and test selection validated. No p-hacking, no multiple comparison errors.\n3. **AI-Pattern Detection** — 41 Humanizer patterns scanned. Density must be \u003C 2 violations per 1000 words.\n4. **Style Audit** — Zero em dashes. Pattern density \u003C 1\u002F2000 words. Voice must match the author's writing profile.\n5. **Adversarial Review** — All 10 reviewer personas must recommend acceptance. Not a subset. All 10.\n\n---\n\n## Live Example: SIREN Paper\n\nThe pipeline was run to produce a full paper on **Intent-Based Blockchain Execution via Agentic RAG and Swarm Consensus**. Complete output in [`examples\u002Fsiren-paper\u002F`](examples\u002Fsiren-paper\u002F):\n\n| File | Description |\n|------|-------------|\n| `siren-paper.pdf` | 13-page compiled paper |\n| `siren-paper.tex` | LaTeX source (504 lines, 26 references) |\n| `figures\u002F*.pdf` | 3 publication-ready figures |\n| `README.md` | Pipeline summary with review scores |\n\n**Review scores progressed from Avg 4.6\u002F10 → 8\u002F10 across 4 revision rounds.**\n\n---\n\n## FAQ\n\n**Q: Does this require a specific LLM provider?**  \nNo. Edit `config\u002Fagent-config.json` to use any OpenAI-compatible or Anthropic API.\n\n**Q: Can I add my own LaTeX template?**  \nYes. Add a folder under `templates\u002F` with `.tex`, `.sty`, and `.cls` files, then update `subagents\u002Fformatter.md`.\n\n**Q: How long does a paper take?**  \n30 minutes to 4 hours depending on LLM speed, literature volume, and revision rounds.\n\n**Q: The output sounds too AI-like. What do I do?**  \nProvide 2-3 paragraphs of your published writing in `data\u002Fvoice-profile\u002F`. The writers will match your voice at the sentence level.\n\n**Q: Can I use this without OpenCode?**  \nThe agents are OpenCode-compatible, but the Python CLI tools (`tools\u002Fliterature_client.py`, `tools\u002Fcitation_verifier.py`) work standalone.\n\n**Q: How do I contribute?**  \nSee [CONTRIBUTING.md](CONTRIBUTING.md). Good first issues are tagged. Template stubs need filling, the PyPI package needs publishing, and more reviewer personas are welcome.\n\n---\n\n## Directory Structure\n\n```\nsisyphus-academica\u002F\n├── orchestrator\u002F          # Research Director agent (the conductor)\n├── subagents\u002F             # Core writing pipeline agents (writer, verifier, etc.)\n├── novelty-engines\u002F       # 6 novelty generation agents\n├── reviewers\u002F             # 10 adversarial reviewer personas\n├── skills\u002F                # 17 portable skill files (also available as standalone [novelty-skills](https:\u002F\u002Fgithub.com\u002Fargahv\u002Fnovelty-skills) repo)\n├── tools\u002F                 # Python CLI toolchain\n│   ├── literature_client.py    # Multi-source lit search\n│   └── citation_verifier.py    # Citation verification + BibTeX\n├── templates\u002F             # LaTeX venue templates (add yours)\n├── config\u002F                # Agent configuration\n├── examples\u002Fsiren-paper\u002F  # Full pipeline output (13-page paper)\n├── data\u002F                  # Research memory + voice profiles\n├── tests\u002F                 # Python unit tests\n├── docs\u002F                  # GitHub Pages documentation\n├── docker-compose.yml     # LaTeX + dev environments\n└── pyproject.toml         # Package metadata\n```\n\n---\n\n## Development\n\n```bash\npip install -r requirements.txt\npython -m pytest tests\u002F -v\nflake8 tools\u002F --max-line-length=100\n\n# LaTeX via Docker\ndocker compose --profile latex run latex pdflatex out\u002Fpapers\u002Fpaper.tex\n```\n\n---\n\n## Acknowledgments\n\n- **[Humanizer](https:\u002F\u002Fgithub.com\u002Fblader\u002Fhumanizer)** by blader — the 30-pattern AI-detection skill this system builds on\n- **[Novelty Skills](https:\u002F\u002Fgithub.com\u002Fargahv\u002Fnovelty-skills)** — standalone thinking tools for AI agents (separate repo)\n- **[OpenCode](https:\u002F\u002Fopencode.ai\u002F)** + **[OhMyOpenAgent](https:\u002F\u002Fomo.dev\u002F)** — agent orchestration platform\n- All six novelty engines were inspired by cognitive diversity research\n\n---\n\n## License\n\nMIT — see [LICENSE](LICENSE) for details.\n\n⭐ **Star this repo if you write research papers — it helps others find it.**\n",2,"2026-06-11 04:12:04","CREATED_QUERY"]