Borrowing it
Nothing to install: this file belongs to Peiiii/nextclaw. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Peiiii/nextclaw/master/.agents/skills/idea-search-funnel/SKILL.mdgit clone --depth 1 https://github.com/Peiiii/nextclawWrote 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/peiiii/nextclaw/idea-search-funnel)<a href="https://agentmods.dev/skills/peiiii/nextclaw/idea-search-funnel"><img src="https://agentmods.dev/badge/skills/peiiii/nextclaw/idea-search-funnel/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/peiiii/nextclaw/idea-search-funnel"><img src="https://agentmods.dev/badge/skills/peiiii/nextclaw/idea-search-funnel.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 | $0.00078 | $0.02081 |
| Opus 5.5 | $0.00031 | $0.00832 |
| Sonnet 5.5 | $0.00016 | $0.00416 |
| Haiku 4.5 | $0.00008 | $0.00208 |
Grade A, and why
idea-search-funnel 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 1mo 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 — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Idea 搜索漏斗
目标
在投入昂贵验证之前,提高“真正值得考虑的方向已经进入候选池”的概率。先把未来判断和搜索空间找全,再批量收集证据和淘汰,最后才把研究成本集中到少数候选;不把第一个像样的想法当作搜索结果,也不把已被市场命名的拥挤品类误当成全部机会空间。
这里优化搜索质量而非数量;同源变体、无因果依据或只换措辞都不算高召回。
搜索合同
开始时冻结并写入探索账本:
- 要寻找什么类型的 idea,以及最终服务哪项决策;
- 时间范围、资源、能力、地域、风险和不可接受损失;
- 什么才算结构不同,哪些只是同一方向的功能或包装变体;
- 当前处于宽搜、批量过滤、半决赛还是决赛;
- 本轮更怕漏掉好方向还是保留太多噪声;
- 何时可以从桌面搜索转入高成本验证。
- 若最终决策要求显著高于行业基准的成功率,什么样的可验证信息差才有资格解释这部分超额概率。
缺失信息不影响广泛搜索时先继续;只有它会改变整个搜索空间时才询问。
可审阅探索账本
高风险或跨轮搜索必须维护一份用户可直接审阅的持久文档,不能只把历程留在对话中。项目已有知识治理时按其规则选择 thought、loop 或其它 owner;没有项目目录时使用单一结构化文档。
账本至少保留:搜索合同与成功定义、覆盖地图、稳定 ID 候选池与机会领地、区分事实/推断/预测/未知的证据、晋级门与历次变化、淘汰墓碑、当前排序和覆盖缺口。顶部提供快照,后部保留探索历程。
更新结论时不得静默改写旧判断;记录证据如何改变排序,方便用户 review。
漏斗阶段
1. 先设计覆盖地图
在生成产品 idea 前,先列出应该搜索的结构维度。至少覆盖:
- 技术能力、成本曲线与基础设施变化;
- 未来会极度丰富的供给与随之产生的稀缺资源;
- 用户行为、工作方式和组织结构变化;
- 产业价值链、利润池和控制点迁移;
- 默认入口、平台激励与生态冲突;
- 分发、网络、品牌、数据、信任和监管;
- 创始人的独特认知、资产、渠道与行动优势;
- 哪些重要未来判断可能真实但尚未被多数竞争者正确定价或行动,以及这种非共识将持续多久;
- 相反世界:主流预测错误时仍可能成立的方向。
先形成稳定 ID 的未来判断池,记录预测、因果链、共识基线、未被正确定价的原因、可观测期限、新稀缺性和捕获者,再推导商业方向。不要把熟悉行业或流行 idea 误当成完整世界。
2. 高召回批量发散
从不同未来力量和稀缺性组合中批量推导机会。高风险 all-in 搜索的原始池通常应达到数十个真正不同的假设,必要时扩展到百级;随后合并为十几个到数十个机会领地。
每个原始候选只记录稳定 ID、未来判断 ID、人群、变化、稀缺性、价值捕获位置和结构差异。只依赖已被定价共识的候选标为基准机会,不能仅靠执行叙事获得超额成功概率。此时不写 PRD,也不为单个候选做真实实验。
分批发散并定期检查新批次是否仍产生新的因果类型。连续批次只出现已有类别的同源变体时,才认为接近覆盖饱和。
3. 按主题批量收集大样本证据
围绕共同假设、用户群、价值链和平台变化收集资料,一份证据应尽量同时更新多个候选,不为每个 idea 重复一套串行研究。
证据优先级为:真实交易、留存与长期行为;大样本和一手行业数据;公开产品行为、定价与平台路线;真实失败;访谈和社区;融资叙事与发布文案。后两类不能单独晋级候选。
记录样本范围、时间、来源偏差和反方证据。资料增加但不再改变候选聚类、排序或置信度时停止堆积。
4. 统一硬过滤
对机会领地批量应用相同过滤器,先淘汰明显不成立的方向:
- 模型能力顺风:基础模型能力提高十倍、成本下降十倍后,产品更强、至少不受损,还是被直接抹平;
- 平台吸收:它是否只是模型、操作系统、工作平台或现有 SaaS 必然内建的单点能力;
- 持久资产:能否积累用户关系、长期状态、专属数据、权限和信任、网络、分发、品牌、监管资质或经济关系;
- 持续价值:未来稀缺性是否对应高频、强烈、可付费且可长期存在的问题;
- 价值捕获:方向即使创造价值,独立产品是否有权收费,还是利润必然被上下游平台拿走;
- 创始人优势:创始人的认知、AI 杠杆、速度、渠道或现有资产能否形成并捕获领先;
- 信息优势:若方向声称成功率显著高于充分竞争市场的基准,是否存在一个具体、重要、尚未被多数竞争者正确定价且可由创始人行动捕获的判断;
- 生存约束:获客、交付、可靠性、信任、法律和资金需求是否适合当前团队。
“今天模型还做不好”不能成为保留理由。在模型能力大陆上,只靠更好提示、单任务推理、暂时集成难度或人工流程自动化建造的产品,默认淘汰;除非它还能积累不随模型升级消失的独立资产。
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 108 lines · 78 tokens per session scan A 7b1687bcc842
idea-search-funnel is a skill published in the GitHub repository Peiiii/nextclaw (263 stars, last pushed yesterday), licensed MIT. It adds 78 tokens to every session and 2,081 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-09-20.
idea-search-funnel compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| idea-search-funnelPeiiii/nextclaw | Skill | 263 | yesterday | MIT | A | 78 | Codex |
| ha-skill-creatorshiwenwen/hope-agent | Skill | 1.9k | yesterday | MIT | A | 106 | Claude Code |
| ha-browsershiwenwen/hope-agent | Skill | 1.9k | yesterday | MIT | A | 140 | Claude Code |
| ha-logsshiwenwen/hope-agent | Skill | 1.9k | yesterday | MIT | A | 183 | Claude Code, Codex |
| ha-pet-importshiwenwen/hope-agent | Skill | 1.9k | yesterday | MIT | A | 109 | Codex |
| ha-skill-installershiwenwen/hope-agent | Skill | 1.9k | yesterday | MIT | A | 82 | Claude Code, Codex |
| email-draftshiwenwen/hope-agent | Skill | 1.9k | yesterday | MIT | A | 46 | Claude Code, Codex |