AI & ML

amart-builder Skill

Session-wide CEO mode - the session model leads a cross-vendor agent team, and the seats are set by whichever model is driving. Fable 5.1 driving - Fable plans, judges, and writes; GPT-5.6 Sol implements in DO mode at high reasoning effort; Claude Opus 5 is the fresh-context adversarial reviewer. Opus 5 driving - Opus implements the code itself; GPT-5.6 Sol is the ADVISE-only planning peer and fresh-context adversarial reviewer; Opus makes every final call. GPT-5.6 Sol driving is the Codex edition of this skill. Cheaper workers (Sonnet 5, Haiku 4.5, GPT-5.6 Terra/Luna, Grok 4.5 for X/Twitter) take bounded work in DO mode (full tools) or ADVISE mode (read-only). Use when the user types /orchestrator or says "orchestrator mode", "delegate mode", "manager mode", or asks for a mode where the smart model manages cheaper agents. Args - "fable" or "opus" force a seat map, "president" keeps the seats with minimal ceremony, "off" ends the mode. Works in Claude Code and in Claude Cowork (desktop app or claude.ai), wher

sbradl Bundle

Deep code-craft review of a diff across five lenses — domain-driven design, code smells, Clean Code, Pragmatic Programmer, and Philosophy of Software Design — each run in its own parallel sub-agent, then aggregated without cross-lens reranking.

zmjjkk123-hub Skill

Forge Codecs 数据序列化：Codec 编码/解码（Codec#encodeStart、Codec#parse）、DynamicOps 动态操作（JsonOps#INSTANCE/COMPRESSED、NbtOps#INSTANCE、#convertTo 格式转换）、DataResult 数据结果（#result、#error、#resultOrPartial）、现有 Codec（BOOL/BYTE/SHORT/INT/LONG/FLOAT/DOUBLE/STRING/BYTE_BUFFER/INT_STREAM/LONG_STREAM/PASSTHROUGH/EMPTY、ResourceLocation#CODEC、CompoundTag#CODEC、Registry#byNameCodec、IForgeRegistry#getCodec）、RecordCodecBuilder 记录 Codec（#create、Instance#group、#fieldOf/#optionalFieldOf、#forGetter、#apply）、Transformer 转换器（#xmap/#flatComapMap/#comapFlatMap/#flatXMap）、Range Codec 范围（#intRange/#floatRange/#doubleRange）、Defaults 默认值（#orElse/#orElseGet）、Unit 单元（Codec#unit）、List Codec（Codec#listOf）、Map Codec（Codec#unboundedMap）、Pair Codec（Codec#pair）、Either Codec（Codec#either）、Dispatch Codec（Codec#dispatch、注册表分派）、MapCodec 扩展、DataFixerUpper 库。

louisfghbvc Bundle

Comprehensive best practices, game loop specifications, prompt architectures, and visual verification workflows derived from Awesome Astra Prompts. Use when creating or iterating on 3D games, playable browser experiences (Three.js/WebGPU), game engine projects (Godot 4, Unity, Unreal, Roblox), flight/vehicle simulators, action RPGs, procedural dioramas, or headless Blender asset pipelines with agy.

zmjjkk123-hub Skill

Minecraft Chicken Variant 鸡变体定义：CHICKEN_VARIANT 注册表、data/<namespace>/chicken_variant/ 数据包路径、JSON 格式（asset_id 成年鸡纹理、baby_asset_id 幼鸡纹理、model 模型类型 normal/cold、spawn_conditions 生成条件）、asset_id/baby_asset_id 纹理解析（assets/<namespace>/textures/<path>.png）、Model 模型类型（normal 温带/热带鸡、cold 寒冷鸡）、Spawn Conditions 生成条件选择器（condition.type 生物群系/月光亮度/结构、biome biomes 标签/ID/列表、moon_brightness range 范围（满月1/新月0）、structure structures 标签/ID/列表、priority 优先级 随机打破平局）、服务器启动加载（/reload 不重新加载）、CHICKEN_VARIANT 注册表至少一个元素、最高优先级有效选择器生成鸡变体。

zmjjkk123-hub Skill

Minecraft Equipment Asset 装备资产定义：Definition Format 定义格式（assets/<namespace>/equipment/ JSON 文件、layers 层级 映射预设模型类型→模型层信息列表）、Layer Fields 层字段（dyeable 染色影响 dyed_color 组件、color_when_undyed 未染色时色调 RGB、texture 纹理 assets/<namespace>/textures/entity/equipment/<preset type>/<path>.png、use_player_texture 翅膀 玩家披风纹理、trim_palette_replacements 盔甲装饰调色板替换 映射 palette ID→replacement palette ID）、Definition Behavior 定义行为（装备槽匹配时渲染预设模型类型层、未定义层不渲染、humanoid 和 wings 可同时渲染）、Preset Model Types 预设模型类型（Humanoid armor 人形盔甲：humanoid 头/胸/脚、humanoid_leggings 腿部、humanoid_baby 婴儿人形 不能渲染装饰、wings 翅膀 抑制披风模型；Animal armor 动物盔甲：wolf_body/horse_body/llama_body/happy_ghast_body/nautilus_body；Saddles 鞍：pig_saddle/strider_saddle/camel_saddle/camel_husk_saddle/horse_saddle/donkey_saddle/mule_saddle/skeleton_horse_saddle/zombie_horse_saddle/nautilus_saddle）。

zmjjkk123-hub Skill

Minecraft Resource Loading 资源加载（MC 1.21.11+ Forge）：Core Rule 核心规则（Resource Location == File Path、namespace:path→src/main/resources/文件、禁止在JSON中包含.json/.png扩展名、命名空间必须精确匹配mods.toml modId和Java @Mod(MODID)、目录名单数固定 items/models/item/models/block/blockstates/textures/item/textures/block/lang/recipe）、Item Model Definition 物品模型定义（assets/<modid>/items/<registry_name>.json 每个物品/方块物品必需 缺失导致库存/搜索图标显示缺失）、Item Model+Texture 物品模型+纹理（assets/<modid>/models/item/<registry_name>.json parent minecraft:item/generated textures layer0、assets/<modid>/textures/item/<registry_name>.png 16x16 PNG）、Block Model+Blockstate+Block Item 方块模型+方块状态+方块物品（assets/<modid>/blockstates/<registry_name>.json variants model、assets/<modid>/models/block/<registry_name>.json parent cube_all、assets/<modid>/models/item/<registry_name>.json parent block、assets/<modid>/textures/block/<registry_name>.png）、Recipe 配方 MC1.21.11+ 格式（ingredients 纯字符串 item id、result 使用 id+count 不是 item+count）、Lang 语言文件（en_us.json/zh_cn.json item/block/itemGroup 翻译键）、pack.mcmeta（min_f

zmjjkk123-hub Skill

Minecraft Model 模型格式：Model Categories 模型类别（Baked model 烘焙模型 几何固定、Dynamic model 动态模型 几何可变；Block model 方块模型、Item model 物品模型、Entity model 实体模型）、Baked Models 烘焙模型文件（assets/<namespace>/models/、JSON parent/ambientocclusion/textures/elements/display/gui_light）、Model Inheritance 模型继承（子模型缺少字段使用父值、display模式继承、textures合并解析、循环继承检测和中断、缺失父模型重定向）、Texture Variables 纹理变量（命名变量代替直接路径、#前缀引用其他变量、缺失→缺失纹理）、Elements 元素（矩形长方体 from/to 端点、rotation 旋转 origin/rescale/axis/angle、shade 着色、light_emission 自发光 0-15、faces 面 texture/uv/rotation/tintindex/cullface）、Display Transforms 显示变换（thirdperson_lefthand/righthand、firstperson_lefthand/righthand、head、gui、ground、fixed、on_shelf；scale/rotation/translation 应用顺序 scale→rotation→translation）、Invalid Model 无效模型 builtin/missing 缺失纹理）、Blockstate Dispatch 方块状态分派（assets/<namespace>/blockstates/、variants 直接映射 prop=value 键、multipart 组合 when条件 AND/OR 属性组）、Candidate Models 候选模型（model/uvlock/x/y/z/weight 选择权重）、Face Culling 面剔除（cullface 方向 occlusion shape 覆盖判断）、Block Particle Texture Variable 方块粒子纹理变量（particl

harikrishnan94 Skill

Reduce unnecessary code and comment lines so reviewers can understand and verify a change without the author's agent history. Use whenever writing or modifying code a human will review — a PR, a patch, a commit on a shared branch — and when cleaning up an existing diff on explicit cues like "humanize this", "this looks AI-generated", "AI slop", "reviewers keep flagging generated code", or "clean this up before I open the PR". Applied to an existing diff it fixes the code by default and reports what changed. Removes verbose comments, planning leftovers, unexplained labels, and redundant code while preserving behavior and useful rationale. Keeps comments, commit messages and PR text in plain sentences. Honors required AI disclosure.

harikrishnan94 Bundle

Use when the user wants a task prompt for an agent to execute — delegated engineering work (build, optimize, investigate, decide, document, repair), especially long-running, unattended, or high-assurance tasks where the result must be trustworthy. Produces a right-sized prompt with a runnable completion oracle, doer≠grader verification, and honest failure paths. NOT for system prompts or chatbot personas, conversational/creative prompts, or wording tweaks to an existing prompt; for tiny tasks it emits a short LIGHT prompt, never the full template.

harikrishnan94 Skill

Produce a self-contained brief the user can paste into a separate, independent model session for a second opinion on a spec, plan, or implementation diff. Use when the user asks for a second opinion, a peer review, an independent review, or wants to hand off the current spec/plan/impl to another model. Mandates severity-labeled findings (Low/Medium/High/Critical) and instructs the reviewer to return its output as a prompt demanding confirm/reject with detailed arguments per finding.

zmjjkk123-hub Skill

Forge 自定义模型加载器：BakedModel 接口（#getOverrides ItemOverrides、#useAmbientOcclusion 环境光遮蔽、#isGui3d GUI 3D 渲染、#isCustomRenderer 自定义渲染器 BEWLR、#getParticleIcon 粒子纹理、#getQuads BakedQuad 列表、#applyTransform 变换）、ItemOverrides 物品覆盖（#resolve 动态模型、#getOverrides BakedOverride 列表）、BakedOverride 物品覆盖匹配（ItemOverrides$PropertyMatcher、predicate 属性匹配）、ItemDisplayContext 显示上下文（NONE/THIRD_PERSON_LEFT_HAND/THIRD_PERSON_RIGHT_HAND/FIRST_PERSON_LEFT_HAND/FIRST_PERSON_RIGHT_HAND/HEAD/GUI/GROUND/FIXED）、ItemTransforms 变换容器、ItemTransform 旋转/平移/缩放、Custom Model Loaders 自定义加载器（IGeometryLoader、IUnbakedGeometry#bake）、WaveFront OBJ 模型（forge:obj 加载器、.obj/.mtl 文件、flip_v V轴翻转）、ModelData 模型数据、ModelProperty 模型属性、CompositeModel$Data 复合模型数据、RenderShape#ENTITYBLOCK_ANIMATED、BlockEntityWithoutLevelRenderer BEWLR 渲染。

harikrishnan94 Bundle

Offload the implementation typing of a settled plan to the Cursor CLI (`cursor-agent`), which reaches Grok, Gemini, GPT/Codex and Composer, while you keep the planning and the verification. Use when the plan is settled and the remaining work is mechanical or wide-but-shallow — scaffolding, a repetitive refactor across many files, filling in tests, a contained fix with a test as its oracle — and cost or throughput matters; also for a read-only review from a different model family. Do NOT use it for ambiguous intent, architecture decisions, security-sensitive changes, or anything under roughly ten minutes of typing, where the per-run startup and the cost of writing the brief exceed what delegation saves.

galiprandi Skill

Autonomous user preference detection, storage and injection. The agent detects preferences from conversation, saves them to Postgres, and loads active ones at the start of every flow.

galiprandi Skill

Configure and adjust the job search strategy level. The agent interrogates the user, proposes a level, and saves it to DB. All flows respect it.

zmjjkk123-hub Skill

Forge 1.21.11 最小可运行 MOD 固定流程：命名规则（modid/Java 包/主类名/物品ID）、固定工作流（activate_test_mode、validate_resources、run_mod_test_cycle）、物品注册（DeferredRegister<Item>、RegistryObject<Item>、Item.Properties.setId、FoodProperties）、物品模型定义（assets/<modid>/items/<name>.json、minecraft:model 类型、models/item/<name>.json）、纹理生成（Python PNG 脚本、16x16 纯色占位符）、语言文件（en_us.json/zh_cn.json、item.<modid>.<name> 键）、配方（data/<modid>/recipe/<name>.json、crafting_shaped、result id/count 格式）、工具快速参考（Item.Properties.sword/pickaxe、ToolMaterial.COPPER、AxeItem）、方块注册（DeferredRegister<Block>、BlockBehaviour.Properties、BlockItem、blockstates/models/block/models/item/items/blockstates.json）、GameTest 测试（@GameTestNamespace、@GameTest、GameTestHelper、Identifier、lookupOrThrow、ResourceKey.create）、复杂功能（自定义盔甲+鞘翅、humanoidArmor、DataComponents.GLIDER、canElytraFly/elytraFlightTick）、常见陷阱（items/<name>.json 缺失、旧配方格式、ArmorItem 类不存在、getModBusGroup/getModEventBus 混淆、ResourceLocation→Identifier、registryOrThrow→lookupOrThrow）、构建/验证纪律（validate_resources、run_mod_test_cycle、run_test_gametest）。

zmjjkk123-hub Skill

Minecraft Armor Trim 盔甲装饰定义：Trim Pattern 装饰图案（TRIM_PATTERN 注册表、data/<namespace>/trim_pattern/、description 文本组件、asset_id 资源 ID、decal 贴花模式）、Trim Material 装饰材料（TRIM_MATERIAL 注册表、data/<namespace>/trim_material/、asset_name 纹理路径后缀、palette_id 调色板 ID、override_armor_assets 资产覆盖）、Tooltips 提示信息（图案名称 + 材料名称显示）、Textures 纹理生成（minecraft:armor_trims 图集、assets/<ns>/textures/trims/entity/<equipment model layer>/<pattern>_<material>.png、paletted_permutations 调色板置换、UV 映射）、Smithing Trim Recipe 锻造装饰配方（provides_trim_material 组件、trim 组件）、Equipment Assets 装备资产（trim_palette_replacements 调色板替换）、服务器启动加载定义。

zmjjkk123-hub Skill

Minecraft Cow Variant 牛变体定义：COW_VARIANT 注册表、data/<namespace>/cow_variant/ 数据包路径、JSON 格式（asset_id 成年牛纹理、baby_asset_id 幼牛纹理、model 模型类型 normal/cold/warm、spawn_conditions 生成条件）、asset_id/baby_asset_id 纹理解析（assets/<namespace>/textures/<path>.png）、Model 模型类型（normal 温带牛、cold 寒冷牛、warm 热带牛）、Spawn Conditions 生成条件选择器（condition.type 生物群系/月光亮度/结构、biome biomes 标签/ID/列表、moon_brightness range 范围（满月1/新月0）、structure structures 标签/ID/列表、priority 优先级 随机打破平局）、服务器启动加载（/reload 不重新加载）、COW_VARIANT 注册表至少一个元素、最高优先级有效选择器生成牛变体。

zmjjkk123-hub Skill

Forge 模型扩展：Face Data 面数据（forge_data 字段、color ARGB 颜色调制、block_light/sky_light 光照覆盖 0-15、ambient_occlusion 环境光遮蔽）、Elements Model 元素模型（元素级/面级数据继承、forge:item_layers 加载器）、Generated Item Model 生成物品模型（纹理层索引映射）、Render Type 渲染类型（render_type 字段、NamedRenderTypeManager、vanilla 值 minecraft:solid/cutout/cutout_mipped/cutout_mipped_all/translucent/tripwire、ForgeRenderTypes 实体渲染类型、RegisterNamedRenderTypesEvent 自定义注册、RenderType.chunkBufferLayers、DefaultVertexFormat.NEW_ENTITY）、Root Transforms 根变换（IGeometryBakingContext#getRootTransform、origin 原点 corner/center/opposing-corner、translation 平移、left_rotation/right_rotation 旋转、scale 缩放、matrix 3x4 矩阵、四元数旋转）、Part Visibility 部件可见性（visibility 字段、composite 模型、OBJ 模型、BakedModel 烘焙）。

zmjjkk123-hub Skill

Minecraft Tutorial: Custom Armor Trims 自定义盔甲装饰教程（Java 1.21.5+）：Data Pack 数据包（Recipe 配方 smithing_trim 模板+基底+添加物+图案、Trim Pattern 装饰图案 asset_id+description+decal、Trim Material 装饰材料 provides_trim_material 数据组件 而非配方字段 asset_name+description）、Resource Pack 资源包（Equipment Model Atlas 装备模型图集 覆盖assets/minecraft/atlases/armor_trims.json 合并非替换原版 paletted_permutations 调色板置换 palette_key+permutations+textures；Item Model Atlas 物品模型图集 覆盖blocks.json 原版纹理复用；Item Models 物品模型 select property minecraft:trim_material 按材料选择层 texture layer0+layer1）。

baraki123 Skill

Save a resume-ready snapshot to context.md §0 so a FRESH agent (no memory of this session) can pick up exactly where this one left off. Use when wrapping up or terminating a Claude Code session, or when the user says "save context", "checkpoint", "hand off", or "I'm closing this session".

zmjjkk123-hub Skill

Minecraft Pig Variant 猪变体定义：PIG_VARIANT 注册表、data/<namespace>/pig_variant/ 数据包路径、JSON 格式（asset_id 成年猪纹理、baby_asset_id 幼猪纹理、model 模型类型 normal/cold、spawn_conditions 生成条件选择器）、asset_id/baby_asset_id 纹理解析（assets/<namespace>/textures/<path>.png）、Model 模型类型（normal 温带/热带猪、cold 寒冷猪）、Spawn Conditions 生成条件选择器（condition.type 生物群系/月光亮度/结构、biome biomes 标签/ID/列表、moon_brightness range 范围（满月1/新月0）、structure structures 标签/ID/列表、priority 优先级 随机打破平局）、服务器启动加载（/reload 不重新加载）、PIG_VARIANT 注册表至少一个元素、最高优先级有效选择器生成猪变体。

rcrdk Skill

Agent: TDD Guide

vickyck428-ux Bundle

Check Amazon US/UK/DE image-search candidates, confirm exact same products with the Volcano vision model, enrich confirmed ASINs with public Amazon and SellerSprite fulfillment/sales/rank data, and calculate data-completeness-safe channel inputs. Use after Google screening or when retrying Amazon FBA/FBM/SellerSprite enrichment.

suisuiwang081-coder Bundle

Canonical entry for taking a short-video product brief, selected creative direction, outline, or story through the approved story-to-video-prompt workflow. When no usable story direction exists, route product-only input through short-video-director-thinking first; when an outline or story already exists, skip direction ideation and refine it directly. Then test duration fit, select shot language when useful, jointly plan timed shots and a compact production bible, and hand off approved locks for AI video-prompt compilation. Use for 无梗概产品创意入口、剧情优化、人物场景细化、分镜结构选择、镜头语言方案、音画分镜与台词定稿，以及从剧情到视频提示词的审核流程. Do not use for reusable Prompt compilation, reference-image boards, or direct media generation.

suisuiwang081-coder Bundle

Stage 5 compiler for turning an approved audiovisual plan, production bible, and optional reference assets into direct one-off AI video-generation prompts, plus local segment, continuity, sound, voice, or stitching repair without rewriting locked creative decisions. Use for TikTok, Reels, Shorts, TikTok Shop, video shot prompts, AI video prompts, model-specific prompt adaptation, model-call segment packaging, continuity, sound/voice guidance, or post-edit stitching. For story creation, duration/structure choice, or plan approval use story-to-reusable-prompt; for storyboard/reference assets use short-video-storyboard-prompts.

suisuiwang081-coder Bundle

TikTok带货短视频方向策划与AI视频导演思考技能。Use when the user provides a product, product link, selling point, audience, or rough idea without a usable story premise, or explicitly asks to compare, replace, or rethink creative directions. Generate and assess product strategy, Seedance-suitable directions, hooks, action chains, product-entry logic, and visual conflicts, then pass only the selected direction through a compact SELECTED_DIRECTION_HANDOFF to story-to-reusable-prompt. This skill chooses directions; it does not write complete stories, dialogue scripts, storyboards, final video prompts, sound plans, or stitching instructions. When a usable outline/story already exists and the user wants it completed or optimized, skip this skill and use story-to-reusable-prompt. 中文触发：带货方向、剧情方向、短剧方向、爆款钩子、商品卖点转剧情、AI视频方向、Seedance方向、产品种草创意、方向初筛。

it235 Skill

Backend implementation skill—API / data model / business logic coding conventions and acceptance checklist; the preferred mount for @BackendDev.

it235 Bundle

查询已部署的 multica 平台 API，获取 workspace / agent 统计信息（agent 运行次数、30 天活动明细、各 workspace 汇总）。当用户需要查 multica 数据（如“每个 workspace 的 agent 运行次数”“workspace 统计”“multica 接口数据”），或访问 <MULTICA_API_URL> 的 API 信息时使用。脚本为纯标准库 Python，无需安装依赖。

pedronishida-suno Skill

Run any question, idea, or decision through a council of 5 AI advisors who independently analyze it, peer-review each other anonymously, and synthesize a final verdict. Based on Karpathy's LLM Council methodology. MANDATORY TRIGGERS: 'council this', 'run the council', 'war room this', 'pressure-test this', 'stress-test this', 'debate this'. STRONG TRIGGERS (use when combined with a real decision or tradeoff): 'should I X or Y', 'which option', 'what would you do', 'is this the right move', 'validate this', 'get multiple perspectives', 'I can't decide', 'I'm torn between'. Do NOT trigger on simple yes/no questions, factual lookups, or casual 'should I' without a meaningful tradeoff (e.g. 'should I use markdown' is not a council question). DO trigger when the user presents a genuine decision with stakes, multiple options, and context that suggests they want it pressure-tested from multiple angles.

pedronishida-suno Bundle

Monitors task execution for skill improvement opportunities. Use during ANY multi-step task, agentic workflow, or work session where the agent uses tools and produces deliverables. Captures patterns, user corrections, workflow insights, and methodology worth preserving as reusable skills. Also triggers in post-task feedback discussions and when the user mentions skill observations, improvements, the observation log, skill taxonomy, or asks the agent to watch for skill opportunities. Also known as "One Skill to Rule Them All" — trigger on this phrase too. IMPORTANT: invoke this skill before the FIRST tool call of any session and before writing or proposing a plan — any turn that will involve a tool call counts, however simple the opener looks. This sentence is the session-start trigger and the only activation layer that survives an unreachable config file; pair it with a CLAUDE.md instruction or a harness session-start hook (references/environments.md) — description matching alone is not enforceable.

pedronishida-suno Bundle

Search and query the personal Obsidian "second brain" vault using a local Ollama model (qwen2.5:7b). Use whenever the user wants to READ, search, ask questions about, or recall anything from their second brain / notes / vault (e.g. "what does my second brain say about X", "search my notes for Y", "read the second brain"). Also exposes a node CRUD CLI for manual create/read/update/delete of vault nodes.

agentx-ai Bundle

Turn an AgentX self-host evaluation into a triaged code fix and a re-run against the same dataset, so a before-and-after comparison means something. Use whenever someone has an evaluation on a local AgentX self-host engine (AgentX-trace-eval, normally http://localhost:4700) and wants to know what to actually change in the code, or wants to re-run an evaluation to compare scores. Also use when a request mentions a self-host evaluation id or dataset id, an agent that scored badly, an AI Analysis or judge findings from the Evaluate tab, or the question "the report tells me what is wrong with the answers but not what to fix in my code". The core move is triaging code-blind judge recommendations against the real source instead of applying them literally.

agentx-ai Bundle

Set an existing Python agent up on AgentX, end to end: read the project API key off the engine, write .env.agentx, install agentx-python from PyPI, initialise the SDK once, instrument the calls that are worth tracing, and then prove it by running the agent and reading its own traces back. Use whenever someone wants to set up, initialise, add tracing to, or monitor their own agent with AgentX; asks to "trace my agent", "instrument this repo", "hook this up to AgentX", "set up AgentX here", or "get my runs into Live Traces"; or has a self-host engine (AgentX-trace-eval, normally http://localhost:4700) and nothing reporting into it yet. Also use when traces are configured but not arriving, since the SDK fails silently by design. The core move is one span where the run begins plus one line of framework auto-instrumentation - not a decorator on every function - and it is not finished until a trace has been fetched back.

agentx-ai Bundle

Turn an AgentX improvement report - production failures a human CONFIRMED during signal review, clustered into issues with recommendations - into triaged fixes applied to the agent's source code. Use whenever someone has an improvement report id from a self-host engine (AgentX-trace-eval, normally http://localhost:4700), asks to "apply the improvement report", "fix what the reviewers confirmed", or mentions the dashboard's Insights > Auto-improve surface. This is the ONLINE counterpart of /eval-fix: the evidence is live traffic a human vouched for, not an offline dataset run, and the core move is the same - triage code-blind recommendations against the real source instead of applying them literally.

x1linwang Skill

Debug a Langflow 1.11.x flow that runs without erroring but produces the wrong result or nothing at all, and run a flow from the command line instead of the Playground. Use this when a Chat Output is empty, a node appears not to have run, an If-Else takes the wrong branch or both outputs look blank, a Prompt Template has grown input fields you did not add, an Agent ignores a tool that is clearly connected, or an API call to /api/v1/run returns 403, gzip garbage, or a partial result. Also use it to pick the right auth scheme for a Langflow API call, since two different schemes apply to different routes.