[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-93140":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":9,"languages":9,"totalLinesOfCode":9,"stars":10,"forks":11,"watchers":12,"openIssues":13,"contributorsCount":13,"subscribersCount":13,"size":13,"stars1d":13,"stars7d":13,"stars30d":14,"stars90d":13,"forks30d":13,"starsTrendScore":13,"compositeScore":15,"rankGlobal":9,"rankLanguage":9,"license":16,"archived":17,"fork":17,"defaultBranch":18,"hasWiki":19,"hasPages":17,"topics":20,"createdAt":9,"pushedAt":9,"updatedAt":21,"readmeContent":22,"aiSummary":23,"trendingCount":13,"starSnapshotCount":13,"syncStatus":24,"lastSyncTime":25,"discoverSource":26},93140,"grill-for-unknowns","nicobailon\u002Fgrill-for-unknowns","nicobailon","Agent skill for finding unknowns, grilling plans, and reaching shared understanding before implementation",null,177,6,126,0,25,45.04,"MIT License",false,"main",true,[],"2026-07-22 04:02:08","# grill-for-unknowns\n\n**An agent skill for finding unknowns, grilling plans, and reaching shared understanding before implementation.**\n\n`grill-for-unknowns` combines docs\u002Fsource-grounded plan interrogation with the “finding your unknowns” strategy: known knowns, known unknowns, unknown knowns, and unknown unknowns.\n\nThe intended behavior is simple:\n\n> Inspect the real territory first, grill the plan and user only where decisions matter, surface blindspots, write down the shared understanding, then build.\n\n## What it does\n\nUse this skill when an agent should **not** rush straight into implementation:\n\n- unfamiliar codebase area,\n- unfamiliar API\u002Flibrary\u002Fplatform docs,\n- vague product\u002Fdesign direction,\n- long-running subagent or coding-agent launch,\n- complex architecture or domain-model decision,\n- failed previous attempt due to assumptions,\n- or any plan that needs to be pressure-tested before building.\n\nThe skill asks the agent to separate:\n\n- **facts** it can discover from docs\u002Fsource\u002Ftests\u002Fconfig,\n- **decisions** it needs the user to make,\n- **domain language** that should be clarified and recorded,\n- and **unknowns** that could materially change implementation.\n\n## Install\n\nThe skill is a standard `SKILL.md` folder, so installation is the same everywhere: copy `plugins\u002Fgrill-for-unknowns` into your agent's skills directory. Clone the repo once:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fnicobailon\u002Fgrill-for-unknowns.git \u002Ftmp\u002Fgrill-for-unknowns\n```\n\n### Claude Code\n\n```bash\nmkdir -p ~\u002F.claude\u002Fskills\ncp -R \u002Ftmp\u002Fgrill-for-unknowns\u002Fplugins\u002Fgrill-for-unknowns ~\u002F.claude\u002Fskills\u002Fgrill-for-unknowns\n```\n\nClaude Code discovers it automatically; invoke it as `\u002Fgrill-for-unknowns` or let it trigger from the description. For a project-scoped install, copy into `.claude\u002Fskills\u002F` inside the repo instead.\n\n### Codex\n\n```bash\nmkdir -p ~\u002F.agents\u002Fskills\ncp -R \u002Ftmp\u002Fgrill-for-unknowns\u002Fplugins\u002Fgrill-for-unknowns ~\u002F.agents\u002Fskills\u002Fgrill-for-unknowns\n```\n\nCodex picks up skills from `~\u002F.agents\u002Fskills` (personal) or `.agents\u002Fskills` inside a repo (project-scoped); invoke with `$grill-for-unknowns` or let implicit matching select it from the description. See the [Codex skills docs](https:\u002F\u002Flearn.chatgpt.com\u002Fdocs\u002Fbuild-skills).\n\n### Hermes Agent\n\n```bash\nmkdir -p ~\u002F.hermes\u002Fskills\u002Fsoftware-development\ncp -R \u002Ftmp\u002Fgrill-for-unknowns\u002Fplugins\u002Fgrill-for-unknowns ~\u002F.hermes\u002Fskills\u002Fsoftware-development\u002Fgrill-for-unknowns\n```\n\nThen start a fresh Hermes session or reload skills, and load it explicitly when needed:\n\n```txt\n\u002Fskill grill-for-unknowns\n```\n\n### Other agents\n\nAny agent that supports `SKILL.md` skill folders can use the same copy-the-folder approach.\n\n## Example prompts\n\n```txt\nUse grill-for-unknowns before implementing this. Inspect the relevant docs\u002Fsource first, then interview me one material question at a time until we have shared understanding.\n```\n\n```txt\nGrill this plan against the codebase and official docs. Do not build yet. Surface unknown unknowns, challenge domain terms, and produce a launch packet.\n```\n\n```txt\nI know what I want when I see it. Use grill-for-unknowns to prototype options and extract my unknown knowns before implementation.\n```\n\n```txt\nBefore spawning subagents, use grill-for-unknowns to turn this fuzzy request into a clear launch packet with assumptions, open questions, verification steps, and rollback risks.\n```\n\n## Package contents\n\n```txt\ngrill-for-unknowns\u002F\n├── README.md\n├── CHANGELOG.md\n├── LICENSE\n├── NOTICE.md\n├── package.json\n└── plugins\u002F\n    └── grill-for-unknowns\u002F\n        ├── SKILL.md\n        ├── README.md\n        ├── LICENSE\n        ├── references\u002F\n        │   ├── upstream-lineage.md\n        │   └── domain-modeling-add-on.md\n        └── templates\u002F\n            ├── ADR.md\n            ├── CONTEXT.md\n            ├── grill-session.md\n            ├── implementation-notes.md\n            └── launch-packet.md\n```\n\n## Credits and lineage\n\nThis skill is substantially inspired by, and should be treated as an adaptation\u002Ffork of, Matt Pocock’s skill composition around `grill-with-docs`:\n\n- Matt Pocock’s `grill-with-docs` skill:  \n  https:\u002F\u002Fgithub.com\u002Fmattpocock\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fengineering\u002Fgrill-with-docs\u002FSKILL.md\n- Matt Pocock’s `domain-modeling` skill:  \n  https:\u002F\u002Fgithub.com\u002Fmattpocock\u002Fskills\u002Ftree\u002Fmain\u002Fskills\u002Fengineering\u002Fdomain-modeling\n- Matt Pocock’s `grilling` skill:  \n  https:\u002F\u002Fgithub.com\u002Fmattpocock\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fproductivity\u002Fgrilling\u002FSKILL.md\n\nAdditional inspiration comes from Thariq’s X article:\n\n- Thariq, “A Field Guide to Fable: Finding Your Unknowns”:  \n  https:\u002F\u002Fx.com\u002Ftrq212\u002Fstatus\u002F2073100352921215386\n\n## How this differs from the upstream skills\n\nThe upstream skills are deliberately minimal and compositional: `grilling` is a ~10-line relentless-interview loop, and `grill-with-docs` composes it with `domain-modeling`. `grill-for-unknowns` keeps that core loop and expands it in five ways:\n\n1. **It covers gaps an interview can't see.** The upstream loop only handles known unknowns — questions someone already knows to ask. The four-quadrant taxonomy adds explicit tactics for unknown knowns (the user's unverbalized taste, extracted through cheap prototypes and contrasting references instead of questions the user can't answer) and unknown unknowns (a blindspot pass over docs\u002Fsource\u002Ftests before the interview starts).\n2. **It defines what a good question is.** \"Relentless\" is a personality, not a quality bar. Here every question must be material, grounded in evidence, and answerable, using a template that forces a citation, a why-it-matters, and a recommended default. Low-risk unknowns become labeled assumptions instead of questions at all.\n3. **It is self-contained.** `grill-with-docs` silently depends on two other skills being installed; an agent that loads only the headline file gets almost nothing. This skill inlines the grilling loop and the domain-modeling rules, so it works dropped into any agent — Hermes, Claude Code, or Codex.\n4. **It extends past the planning boundary.** Upstream ends at \"don't build until shared understanding is confirmed.\" This skill covers what happens after alignment: a launch packet for subagents and long-running coding agents, a deviation policy (continue-and-log vs. stop-and-ask), implementation notes for newly discovered unknowns, and a post-implementation explainer\u002Fquiz.\n5. **It produces durable artifacts.** Session ledgers, launch packets, implementation notes, `CONTEXT.md`, and ADRs give the shared understanding a reusable written shape instead of leaving it in the chat transcript.\n\n## License\n\nMIT. See [`LICENSE`](LICENSE).\n\nThe upstream `mattpocock\u002Fskills` repository is MIT licensed with copyright held by Matt Pocock. Because this skill is an adaptation\u002Ffork rather than a totally original work, the license preserves Matt Pocock’s MIT copyright notice and adds Nico Bailon’s copyright notice for this adaptation.\n","这是一个面向AI编程代理的协作式风险识别技能，用于在实施前系统性地识别知识盲区、验证计划假设并达成共识。它基于‘未知的未知’分类框架（known knowns\u002Funknowns等），结合代码库、文档和测试的实证分析，自动区分可验证事实、需用户决策项、待澄清领域术语及潜在影响实现的关键未知项。适用于代码理解初期、API集成、需求模糊场景、架构设计评审或历史失败复盘等需要前置压力测试的开发环节。",2,"2026-07-12 02:30:02","CREATED_QUERY"]