Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add AutoResearch-Factory/Agon/plugin install agonWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/autoresearch-factory/agon/sibyl)<a href="https://agentmods.dev/skills/autoresearch-factory/agon/sibyl"><img src="https://agentmods.dev/badge/skills/autoresearch-factory/agon/sibyl/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/autoresearch-factory/agon/sibyl"><img src="https://agentmods.dev/badge/skills/autoresearch-factory/agon/sibyl.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00043 | $0.01465 |
| Opus 5 | $0.00022 | $0.00732 |
| Sonnet 5 | $0.00009 | $0.00293 |
| Haiku 4.5 | $0.00004 | $0.00146 |
Grade A, and why
sibyl scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 24 lines — stays where its author put it; the contents beside it link to each section on GitHub.
- skills_sibyl/sibyl-ideation: 研究构思三角——理论推导 (theoretical)、跨界移植 (innovator)、远场映射 (interdisciplinary). 生成/审视研究 idea 时用. Research ideation triad — proof-driven, cross-pollination, distant-field mapping. Use when generating or scrutinizing research ideas.
- skills_sibyl/sibyl-debate: 研究辩论五角——魔鬼代言人 (contrarian)/信号提取者 (optimist)/统计怀疑者 (skeptic)/景观定位者 (comparativist)/战略顾问 (strategist). 多视角评估实验结果或论点时用. Research debate quintet — devil's advocate, evidence-backed optimist, statistical skeptic, landscape comparativist, strategic advisor. Use when evaluating experimental results or arguments from multiple angles.
- skills_sibyl/sibyl-team-meeting: 主持人驱动的多角色研究会议协议——agenda、固定角色、2-3轮发言、逐轮总结、共识/分歧/最终决策. 复杂研究决策、实验路线选择、proposal/paper strategy 需要结构化收敛时用；若只需快速多视角审视, 优先用 sibyl-debate. Chair-led multi-role research meeting protocol — agenda, roles, rounds, round summaries, consensus/disagreement, and final decision. Use when a hard research decision needs structured convergence rather than quick lenses.
- skills_sibyl/sibyl-methodology: 实验方法学三角——证伪优先 (empiricist)/实验设计 (planner)/方法审计 (methodologist). 设计/审计实验方案时用. Experimental methodology triad — falsification-first, experiment design, method audit. Use when designing or auditing experiment plans.
- skills_sibyl/sibyl-critique: 论文审查三角——缺陷分类学 (critic)/就绪标准 (final_critic)/分段审查 (section_critic). 全方位审查论文时用. Paper critique triad — flaw taxonomy, readiness criteria, section-level review. Use when reviewing a paper holistically.
- skills_sibyl/sibyl-judgment: 研究判断——回溯推理 (revisionist) + 工程现实主义 (pragmatist). 重新评估假设/评估工程可行性时用. Research judgment — backwards reasoning from data + engineering realism. Use when reassessing hypotheses or evaluating engineering feasibility.
- skills_sibyl/sibyl-writing-craft: 写作技艺——视觉沟通/notation+glossary 统一/顺序写作纪律/per-section 要求/跨 section 一致性核查. 撰写/整合论文草稿时用. Writing craft — visual communication, notation/glossary unification, sequential discipline, per-section requirements, cross-section consistency. Use when drafting or integrating a paper manuscript.
- skills_sibyl/sibyl-latex: LaTeX 排版——模板纪律/tabular 规范/图片规范/编译纪律. 排版/编译论文时用. LaTeX typesetting — template discipline, table rules, figure rules, compilation discipline. Use when typesetting or compiling a paper.
- skills_sibyl/sibyl-experiments: 实验执行——代码质量/pilot first/错误诊断/drift 检测/干预触发/弹性/resource sharing/文件隔离/checkpointing. 在远程 GPU 上执行实验时用. Experiment execution — code quality, pilot-first, error patterns, drift detection, intervention, resilience, resource sharing, file isolation, checkpointing. Use when running experiments on remote GPUs.
- skills_sibyl/sibyl-gates: 质量门与决策——NeurIPS 校准评分/实验 PROCEED-PIVOT/idea ADVANCE-REFINE-PIVOT 决策矩阵/第三方审查/审稿人仿真. 评估研究质量或做出阶段决策时用. Quality gates and decisions — NeurIPS-calibrated scoring, experiment PROCEED/PIVOT, idea validation decision matrix, third-party review, reviewer simulation. Use when evaluating research quality or making stage-gate decisions.
- skills_sibyl/sibyl-landscape: 研究景观——新颖性碰撞分类/scoring/反模式 + 文献调研原则/实施策略 (Adopt-Extend-Compose-Build). 验证 idea 新颖性或做文献调研时用. Research landscape — novelty collision classification, scoring, anti-patterns + literature survey principles, implementation strategy. Use when validating idea novelty or conducting literature surveys.
- skills_sibyl/sibyl-rebuttal: 审稿回复——9 角色协作: concern 分解/回复策略/证据收集 (scholar+theorist+experiment)/写作/creative advocacy/QA 审查. 撰写/审查审稿回复时用. Rebuttal — 9-agent collaborative: concern decomposition, response strategies, evidence gathering, writing, creative advocacy, QA review. Use when writing or reviewing a rebuttal.
- skills_sibyl/sibyl-common: 通用约定——模型选择与时间预算/远程服务器纪律/自我进化安全 (test→commit→push)/质量标准. 任何时候都适用. Common conventions — model selection, time budget, remote server discipline, self-evolution safety, quality standards. Always applicable.
- skills_sibyl/sibyl-reflection: 反思与综合——8 类 issue 分类学/fix tracking/好建议 vs 坏建议/synthesis 原则 (非妥协)/result debate synthesis. 迭代反思/综合多视角/制定改进计划时用. Reflection and synthesis — 8-category issue taxonomy, fix tracking, good vs bad recommendations, synthesis principles, result debate synthesis. Use when reflecting on an iteration, synthesizing perspectives, or creating improvement plans.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 24 lines · 43 tokens per session scan A 22586700e8f3
sibyl is a skill published in the GitHub repository AutoResearch-Factory/Agon (48 stars, last pushed 6d ago), licensed MIT. It adds 43 tokens to every session and 1,465 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
assimilate-popular-workflows
This skill should be used when the user asks to "find skills in the wild", "assimilate popular workflows", "discover SKILL.md files in repos", "research external skills", "find workflow patterns", "survey the skill landscape", "what skills exist out there", or wants to investigate public repositories for extractable…
mcp-app-verification
Comprehensive verification checklists for MCP Apps. Tests with basic-host reference, validates handler-before-connect, text fallback, resource URI linking, single-file bundling, host styling, CSP, and legacy pattern detection.
guardrails-ai-setup
Guardrails AI validation framework setup for LLM applications. Implement input/output validation, safety checks, and structured output enforcement.
cog-braindump-capture
Capture raw thoughts with automatic domain classification and vault routing.
quality-hooks
Language-specific auto-lint/format/typecheck pipeline. Supports Python (ruff+pyright), TypeScript (prettier+eslint+tsc), Go (gofmt+golangci-lint). Auto-fix and convergence loops.
cog-knowledge-consolidation
Build structured knowledge frameworks from scattered vault notes with source attribution.