[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MZ2RDZZ888PPNSNSE64AVX":3,"skill-same:01M6MZ2RDZZ888PPNSNSE64AVX":66},{"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":39,"linkOnly":40,"markdownContent":41,"markdownBody":42,"frontMatter":43,"refName":44,"commitSha":45,"skillDirectory":46,"forksCount":47,"repositoryHtmlUrl":48,"repositoryHomepageUrl":6,"ownerHtmlUrl":49,"ownerBlogUrl":6,"ownerDescription":50,"updatedAt":51,"install":52,"resources":65},"01M6MZ2RDZZ888PPNSNSE64AVX","llm-output-safety",null,"Production safety requirements for pipelines that put LLM-generated text into client-facing outputs: non-Latin character sanitization, programmatic citation integrity, and treating recurring issues as architectural deficiencies. Use when generating reports, exports, or any client-facing document from LLM output.","snewhouse\u002Faa-ma-forge","snewhouse","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6317143?v=4","01M6MYYWF28C4KBQHHWB9SC6TS","Stephen J Newhouse","01M6MYYWJWQMDN2ZPFMJ584SSM",6,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fsnewhouse\u002Faa-ma-forge\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fllm-output-safety\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"llm output safety","Safety","output","document","for","sanitization","programmatic","client","as","requirements","character","issues","reports","pipelines","use","recurring","generated","text","integrity","Production","2026-11-10 10:12:30",false,"---\nname: llm-output-safety\ndescription: \"Production safety requirements for pipelines that put LLM-generated text into client-facing outputs: non-Latin character sanitization, programmatic citation integrity, and treating recurring issues as architectural deficiencies. Use when generating reports, exports, or any client-facing document from LLM output.\"\n---\n\n# LLM Output Safety Rules\n\nDerived from lessons L-050, L-051, L-052. These are production safety requirements for any pipeline that uses LLM-generated text in client-facing outputs.\n\n## Non-Latin Character Sanitization (L-050)\n\n**Trigger:** Any LLM-generated text destined for English-language reports or exports.\n\n**Requirements:**\n- Detect and flag non-Latin script characters (CJK, Cyrillic, Arabic, etc.) in English outputs\n- Strip or replace with appropriate English text\n- Log every sanitization event for audit trail\n- Run at the boundary between LLM response generation and report assembly\n- Integration point: the post-processing method that is the universal boundary across all analyzers\n\n**Why:** LLM training data leaks (e.g., Chinese character \"犬\" appearing in an English veterinary report) destroy ALL credibility.\n\n## Citation Integrity (L-051)\n\n**Requirements:**\n- Citation assignment MUST be **programmatic and deterministic**, not LLM-dependent\n- Post-generation validation: verify cited reference IDs exist in the reference list\n- Cardinality rules: flag >5 citations per paragraph as suspicious, >10 as certain hallucination\n- LLMs are fundamentally unreliable at citation attribution — this must be a system-level guarantee\n\n## Recurring Issues = Architectural Deficiency (L-052)\n\n**Rule:** When the same CLASS of issue appears in 2+ independent QA reviews, treat it as an **architectural deficiency** requiring structural redesign, not a per-report bug fix.\n\n**Required architectural responses:**\n- Timeout resilience: critical systems must have retry logic with extended timeouts\n- Citation pipeline: fully programmatic with deterministic numbering\n- Output validation: LLM text must pass validation before export (non-ASCII, citation patterns, consistency)\n- Executive summary cross-validation: programmatically verify claims match structured analysis data\n","# LLM Output Safety Rules\n\nDerived from lessons L-050, L-051, L-052. These are production safety requirements for any pipeline that uses LLM-generated text in client-facing outputs.\n\n## Non-Latin Character Sanitization (L-050)\n\n**Trigger:** Any LLM-generated text destined for English-language reports or exports.\n\n**Requirements:**\n- Detect and flag non-Latin script characters (CJK, Cyrillic, Arabic, etc.) in English outputs\n- Strip or replace with appropriate English text\n- Log every sanitization event for audit trail\n- Run at the boundary between LLM response generation and report assembly\n- Integration point: the post-processing method that is the universal boundary across all analyzers\n\n**Why:** LLM training data leaks (e.g., Chinese character \"犬\" appearing in an English veterinary report) destroy ALL credibility.\n\n## Citation Integrity (L-051)\n\n**Requirements:**\n- Citation assignment MUST be **programmatic and deterministic**, not LLM-dependent\n- Post-generation validation: verify cited reference IDs exist in the reference list\n- Cardinality rules: flag >5 citations per paragraph as suspicious, >10 as certain hallucination\n- LLMs are fundamentally unreliable at citation attribution — this must be a system-level guarantee\n\n## Recurring Issues = Architectural Deficiency (L-052)\n\n**Rule:** When the same CLASS of issue appears in 2+ independent QA reviews, treat it as an **architectural deficiency** requiring structural redesign, not a per-report bug fix.\n\n**Required architectural responses:**\n- Timeout resilience: critical systems must have retry logic with extended timeouts\n- Citation pipeline: fully programmatic with deterministic numbering\n- Output validation: LLM text must pass validation before export (non-ASCII, citation patterns, consistency)\n- Executive summary cross-validation: programmatically verify claims match structured analysis data\n",{"name":5,"description":7},"main","cd9a2962c9a7e0797fe440f7b1ba4ceccc16ae0d","claude-code\u002Fskills\u002Fllm-output-safety",1,"https:\u002F\u002Fgithub.com\u002Fsnewhouse\u002Faa-ma-forge","https:\u002F\u002Fgithub.com\u002Fsnewhouse","Data Scientist, Bioinformatician, Dad, Husband, MS warrior","2026-11-12 06:45:42",{"source":8,"skillName":5,"commands":53,"prompt":63,"zipAvailable":64},[54,57,60],{"tool":55,"command":56},"npx","npx skills add snewhouse\u002Faa-ma-forge --skill llm-output-safety",{"tool":58,"command":59},"bunx","bunx skills add snewhouse\u002Faa-ma-forge --skill llm-output-safety",{"tool":61,"command":62},"pnpm","pnpm dlx skills add snewhouse\u002Faa-ma-forge --skill llm-output-safety","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fsnewhouse\u002Faa-ma-forge\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fllm-output-safety\u002FSKILL.md",true,[],[67,91,118,139,164,194],{"skillNo":68,"name":69,"nameZh":6,"description":70,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":71,"categories":72,"keywords":78,"publishedAt":39},"01M6MZ588SD5DDH59H18PRNWEV","deploy-helper","You can use this to help with deploy stuff, deploying, deployments, releases, shipping, pushing to prod, going live, rollouts.","https:\u002F\u002Fgithub.com\u002Fsnewhouse\u002Faa-ma-forge\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fevals\u002Fskill-draft\u002Fdraft\u002FSKILL.md",[73],{"categoryNo":74,"categoryName":75,"taxonomyType":76,"taxonomyTypeDesc":77},"01JBSEEDCAT000000000000011","运维与部署","SCENARIO","使用场景",[79,80,81,82,83,84,85,86,87,33,88,89,90],"deploy","deploy helper","helper","you","live","with","can","rollouts","shipping","deployments","releases","help",{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":95,"licenseFileUrl":96,"sourceHtmlUrl":97,"categories":98,"keywords":110,"publishedAt":39},"01M6MZ54NEKRYXGMJ1KJWYMKM7","writing-for-agents","Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.","MIT","https:\u002F\u002Fgithub.com\u002Fsnewhouse\u002Faa-ma-forge\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fwriting-for-agents\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fsnewhouse\u002Faa-ma-forge\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fwriting-for-agents\u002FSKILL.md",[99,102,105],{"categoryNo":100,"categoryName":101,"taxonomyType":76,"taxonomyTypeDesc":77},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":103,"categoryName":104,"taxonomyType":76,"taxonomyTypeDesc":77},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":106,"categoryName":107,"taxonomyType":108,"taxonomyTypeDesc":109},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[111,112,23,113,33,114,115,116,117],"writing","writing for agents","line","instructions","read","editing","documents",{"skillNo":119,"name":120,"nameZh":6,"description":121,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":122,"categories":123,"keywords":124,"publishedAt":39},"01M6MZ500T0FH1G8TBSQV7G5G2","verify-plan","Slash entry for plan verification: runs plan-verification on an existing AA-MA plan, writes [task]-verification.md, and with --iterate loops until the plan converges. Use when the user types \u002Fverify-plan or wants the verification report saved.","https:\u002F\u002Fgithub.com\u002Fsnewhouse\u002Faa-ma-forge\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fverify-plan\u002FSKILL.md",[],[125,126,127,128,129,33,130,84,131,132,133,134,135,136,137,23,138],"verify","plan","verify plan","loops","task","user","verification","report","types","plan verification","entry","iterate","Existing","saved",{"skillNo":140,"name":141,"nameZh":6,"description":142,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":143,"categories":144,"keywords":145,"publishedAt":39},"01M6MZ4V9ST4K54NT1SRBGCE7P","verify-impl","Post-impl adversarial review symmetric to \u002Fverify-plan. Dispatches up to 5 parallel audit agents (code-reviewer, security-auditor, tdd-sequence-auditor, context7-evidence-auditor, future-proofing-auditor) based on the milestone's plan-declared Audit-Profile. CRITICAL findings surface via AskUserQuestion accept\u002Fdispute\u002Fdefer panel before §7.3 user authorization. Introduced in v0.8.0 per ADR-0005. Invoked by Phase 6.8 of \u002Fexecute-aa-ma-milestone.","https:\u002F\u002Fgithub.com\u002Fsnewhouse\u002Faa-ma-forge\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Fverify-impl\u002FSKILL.md",[],[146,147,125,148,149,150,151,152,153,154,155,156,157,158,159,126,160,161,162,163],"impl","verify impl","adversarial","post","panel","via","critical","review","context7","up","evidence","reviewer","parallel","adr","audit","code","tdd","milestone",{"skillNo":165,"name":166,"nameZh":6,"description":167,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":168,"categories":169,"keywords":174,"publishedAt":39},"01M6MZ4K7CC9PMJ1KEY3F53RS1","understand-codebase","Onboard to a new, inherited, or shared codebase. Read it, understand it, map it, and learn its conventions, versioning, tests, tech stack, rules (AGENTS.md \u002F CLAUDE.md \u002F .cursorrules \u002F etc.), build\u002Frun\u002FCI, integrations, security posture, and repo health — then produce an honest pros\u002Fcons verdict, a \"contribute safely\" playbook, and an \"add a feature\" playbook, written to ONBOARDING.md at the repo root plus a .claude\u002Fonboarding\u002F set of deep-dives — and optionally author an AGENTS.md if one is missing, or review and propose improvements to an existing one. Tiered (Quick \u002F Standard \u002F Deep). Reuses codemem (index + diagrams), gsd-map-codebase, system-mapping, code-intelligence, impact-analysis rather than re-implementing them; Deep tier runs a TeamCreate agent-team. Keywords: new codebase, shared codebase, inherited code, onboard, understand this repo, how do I contribute, how do I add a feature, ramp up, get oriented, ONBOARDING.md, AGENTS.md, codebase walkthrough, joining a project.","https:\u002F\u002Fgithub.com\u002Fsnewhouse\u002Faa-ma-forge\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Funderstand-codebase\u002FSKILL.md",[170,171],{"categoryNo":103,"categoryName":104,"taxonomyType":76,"taxonomyTypeDesc":77},{"categoryNo":172,"categoryName":173,"taxonomyType":108,"taxonomyTypeDesc":109},"01JBSEEDCAT000000000000038","人力资源",[175,176,177,155,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193],"understand codebase","codebase","understand","walkthrough","analysis","SET","ci","learn","diagrams","repo health","implementing","re","quick","impact","get","project","deep","shared","map",{"skillNo":195,"name":196,"nameZh":6,"description":197,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":198,"categories":199,"keywords":200,"publishedAt":39},"01M6MZ4D1ZC8MTW064QXY6WC76","token-compression","Output token compression for AA-MA workflows. Reduces response tokens by 40-75% while preserving full technical accuracy. Three intensity levels mapped to execution modes: lite (HITL), full (general), ultra (AFK). Auto-activates during AA-MA execution based on task Mode field. Manual activation via \u002Fcompress lite|full|ultra.","https:\u002F\u002Fgithub.com\u002Fsnewhouse\u002Faa-ma-forge\u002Fblob\u002Fmain\u002Fclaude-code\u002Fskills\u002Ftoken-compression\u002FSKILL.md",[],[201,202,203,204,21,205,206,23,207,208,151,209,210,211,212,129,213,214,215,216],"compression","token compression","token","workflows","modes","accuracy","mode","manual","intensity","tokens","hitl","three","general","auto","technical","execution"]