[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7BTEGZZ0ZSDF73XJMA2ADFN":3,"skill-same:01M7BTEGZZ0ZSDF73XJMA2ADFN":51},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":6,"ownerDescription":29,"updatedAt":30,"install":31,"resources":44},"01M7BTEGZZ0ZSDF73XJMA2ADFN","prompt-contract",null,"Fill the nine-slot prompt contract, then grill, then hand off to a clean Claude session. Use for new substantial tasks. Do not implement in this session.","sardul3\u002Fdev-conductor","sardul3","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F15206322?v=4","01M7BSB2NHRMBNC64JV4MS68FW","Sagar Poudel","01M7BTDJ7EB5S0711S7VCKYVX1",0,"MIT","https:\u002F\u002Fgithub.com\u002Fsardul3\u002Fdev-conductor\u002Fblob\u002Fmain\u002Fskills\u002Fprompt-contract\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: prompt-contract\ndescription: Fill the nine-slot prompt contract, then grill, then hand off to a clean Claude session. Use for new substantial tasks. Do not implement in this session.\n---\n\n# Prompt contract\n\nYou are in the **enrichment** session. Do **not** implement the user's task here. Do not open extra Cursor tabs. The work happens in a new `claude` CLI terminal after launch.\n\nIf the user typed `\u002Fdeep-ask ...`, treat the rest of the message as the original request.\n\n## 1. Fill these nine slots\n\nInfer when obvious. Ask only for missing **high-impact** slots. Multiselect is allowed for role, audience, and length\u002Ftone. Do not ask for a raw model id unless they volunteer an override (`opus`, `sonnet`, `ultra`, or a profile name).\n\n| # | Slot | Notes |\n| - | ---- | ----- |\n| 1 | Role | senior backend, staff SRE, tech lead, security reviewer, … |\n| 2 | Audience | implementing agent, PR reviewer, beginner, future-you |\n| 3 | Goal | one-sentence deep ask **and** the deliverable |\n| 4 | Context | repo, stack, files, constraints; or “none \u002F infer from repo” |\n| 5 | Constraints \u002F non-goals | must \u002F must not |\n| 6 | Output format | code+tests, plan only, diff, ADR, commands |\n| 7 | Length \u002F tone | short, informal, rigorous, production-ready |\n| 8 | Success criteria | observable done-when |\n| 9 | Model profile | `fast` `code` `reason` `heavy` `vision` (auto; user may override) |\n\nProfile tie-break: vision if screenshots\u002FUI; else heavy if many files \u002F production \u002F security; else reason if debug\u002Fdesign\u002Finvestigate; else code if the output is code\u002Ftests; else fast.\n\nOptional (only if they change the work): examples, tools\u002Fsources, plan-then-code vs jump-to-code.\n\nSee `reference.md` in this skill for the assembled markdown shape.\n\n## 2. Grill\n\nFollow `~\u002F.claude\u002Fskills\u002Fdesign-tree-interview\u002FSKILL.md` (or `\u002Fgrill-plan`). Fill remaining contract slots from the tree. Escape: “just go”, “skip grill”, `\u002Fskip-enrich`. If they get lost, `restate-plain`.\n\n## 3. Handoff and launch\n\nWhen the frontier is empty (or they skip):\n\n1. Follow `~\u002F.claude\u002Fskills\u002Fsession-handoff\u002FSKILL.md`. Start the file with `\u003C!-- PROMPT_CONTRACT_V1 -->` and include the nine slots plus original request.\n2. Run (replace the path with the file you wrote):\n\n```bash\n~\u002F.claude\u002Fhooks\u002Fprompt-enrich\u002Flaunch-clean-claude.sh --file \u003Chandoff-path>\n```\n\n3. Tell the user the saved path and that **work continues in the new terminal**. Suggest that session load `implement-terse`.\n4. Stop. Do not start coding in this session. The new tab is the only implementer; this tab continuing will 429 OpenRouter.\n","# Prompt contract\n\nYou are in the **enrichment** session. Do **not** implement the user's task here. Do not open extra Cursor tabs. The work happens in a new `claude` CLI terminal after launch.\n\nIf the user typed `\u002Fdeep-ask ...`, treat the rest of the message as the original request.\n\n## 1. Fill these nine slots\n\nInfer when obvious. Ask only for missing **high-impact** slots. Multiselect is allowed for role, audience, and length\u002Ftone. Do not ask for a raw model id unless they volunteer an override (`opus`, `sonnet`, `ultra`, or a profile name).\n\n| # | Slot | Notes |\n| - | ---- | ----- |\n| 1 | Role | senior backend, staff SRE, tech lead, security reviewer, … |\n| 2 | Audience | implementing agent, PR reviewer, beginner, future-you |\n| 3 | Goal | one-sentence deep ask **and** the deliverable |\n| 4 | Context | repo, stack, files, constraints; or “none \u002F infer from repo” |\n| 5 | Constraints \u002F non-goals | must \u002F must not |\n| 6 | Output format | code+tests, plan only, diff, ADR, commands |\n| 7 | Length \u002F tone | short, informal, rigorous, production-ready |\n| 8 | Success criteria | observable done-when |\n| 9 | Model profile | `fast` `code` `reason` `heavy` `vision` (auto; user may override) |\n\nProfile tie-break: vision if screenshots\u002FUI; else heavy if many files \u002F production \u002F security; else reason if debug\u002Fdesign\u002Finvestigate; else code if the output is code\u002Ftests; else fast.\n\nOptional (only if they change the work): examples, tools\u002Fsources, plan-then-code vs jump-to-code.\n\nSee `reference.md` in this skill for the assembled markdown shape.\n\n## 2. Grill\n\nFollow `~\u002F.claude\u002Fskills\u002Fdesign-tree-interview\u002FSKILL.md` (or `\u002Fgrill-plan`). Fill remaining contract slots from the tree. Escape: “just go”, “skip grill”, `\u002Fskip-enrich`. If they get lost, `restate-plain`.\n\n## 3. Handoff and launch\n\nWhen the frontier is empty (or they skip):\n\n1. Follow `~\u002F.claude\u002Fskills\u002Fsession-handoff\u002FSKILL.md`. Start the file with `\u003C!-- PROMPT_CONTRACT_V1 -->` and include the nine slots plus original request.\n2. Run (replace the path with the file you wrote):\n\n```bash\n~\u002F.claude\u002Fhooks\u002Fprompt-enrich\u002Flaunch-clean-claude.sh --file \u003Chandoff-path>\n```\n\n3. Tell the user the saved path and that **work continues in the new terminal**. Suggest that session load `implement-terse`.\n4. Stop. Do not start coding in this session. The new tab is the only implementer; this tab continuing will 429 OpenRouter.\n",{"name":5,"description":7},"main","e34b29bf2163fe2ec07cc48ea9968ba527ad12b8","skills\u002Fprompt-contract","https:\u002F\u002Fgithub.com\u002Fsardul3\u002Fdev-conductor","https:\u002F\u002Fgithub.com\u002Fsardul3","Software Engineer","2026-09-02 22:42:21",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add sardul3\u002Fdev-conductor --skill prompt-contract",{"tool":37,"command":38},"bunx","bunx skills add sardul3\u002Fdev-conductor --skill prompt-contract",{"tool":40,"command":41},"pnpm","pnpm dlx skills add sardul3\u002Fdev-conductor --skill prompt-contract","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fsardul3\u002Fdev-conductor\u002Fblob\u002Fmain\u002Fskills\u002Fprompt-contract\u002FSKILL.md",true,[45],{"path":46,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":49,"sourceHtmlUrl":50},"reference.md","REFERENCE","参考文档",650,"https:\u002F\u002Fgithub.com\u002Fsardul3\u002Fdev-conductor\u002Fblob\u002Fmain\u002Fskills\u002Fprompt-contract\u002Freference.md",[52,59,66,73,80,87],{"skillNo":53,"name":54,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":19},"01M7BTF29TKRG4X6ZEN51MAE44","verify-before-done","Use before claiming work is complete, fixed, or passing, and before commit or PR. Run the proof command in this turn.","https:\u002F\u002Fgithub.com\u002Fsardul3\u002Fdev-conductor\u002Fblob\u002Fmain\u002Fskills\u002Fverify-before-done\u002FSKILL.md",[],[],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":19},"01M7BTF0ESZZ0NCF01DWXYA7HD","token-playbook","Token budget for Claude Code. Use when shrinking CLAUDE.md, choosing skills vs rules vs MCP, compressing tool output, or cutting always-on context.","https:\u002F\u002Fgithub.com\u002Fsardul3\u002Fdev-conductor\u002Fblob\u002Fmain\u002Fskills\u002Ftoken-playbook\u002FSKILL.md",[],[],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":19},"01M7BTEYMR6SZ108SFAG2ZZ3G5","test-writer","Write failing tests from a technical spec and public contracts only. Use in \u002Fdev-loop test-writer stage. Do not read implementation sources or write production code.","https:\u002F\u002Fgithub.com\u002Fsardul3\u002Fdev-conductor\u002Fblob\u002Fmain\u002Fskills\u002Ftest-writer\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":19},"01M7BTEVAZ353YZ6E9D7HEG80E","tdd","Use when implementing a feature or bugfix, before writing production code. Test first, watch it fail, then implement. Vertical slices at public seams.","https:\u002F\u002Fgithub.com\u002Fsardul3\u002Fdev-conductor\u002Fblob\u002Fmain\u002Fskills\u002Ftdd\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":19},"01M7BTESN9S6ZXSVF7YQJB94R8","systematic-debugging","Use when hitting a bug, test failure, or unexpected behavior, before proposing a fix. Find root cause first.","https:\u002F\u002Fgithub.com\u002Fsardul3\u002Fdev-conductor\u002Fblob\u002Fmain\u002Fskills\u002Fsystematic-debugging\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":19},"01M7BTER6EM870R5PJF6SBBVTH","story-spec","Turn a Jira story into a technical spec via grill. Use in the spec stage of \u002Fdev-loop. Do not implement.","https:\u002F\u002Fgithub.com\u002Fsardul3\u002Fdev-conductor\u002Fblob\u002Fmain\u002Fskills\u002Fstory-spec\u002FSKILL.md",[],[]]