[{"data":1,"prerenderedAt":432},["ShallowReactive",2],{"content-\u002Fplugins\u002Faio-claude-toolkit\u002Faio-skillify":3,"children-\u002Fplugins\u002Faio-claude-toolkit\u002Faio-skillify":431,"twin-\u002Fplugins\u002Faio-claude-toolkit\u002Faio-skillify":426},{"id":4,"title":5,"author":6,"body":7,"budget_tier":6,"build_tags":6,"created":6,"description":422,"document_type":423,"extension":424,"game":6,"install":36,"investment_tier":6,"league":6,"meta":425,"navigation":426,"patch":6,"path":427,"plugin":24,"profit_per_hour":6,"ratings":6,"seo":428,"skills_count":6,"status":6,"stem":429,"strategy_tier":6,"tags":6,"updated":6,"version":6,"weight":6,"__hash__":430},"content\u002Fplugins\u002Faio-claude-toolkit\u002Faio-skillify.md","aio-skillify",null,{"type":8,"value":9,"toc":410},"minimark",[10,37,42,45,50,53,57,62,73,76,80,83,108,112,115,120,128,133,185,191,211,214,220,225,233,236,240,243,253,259,306,311,325,330,374,378,381,393,396],[11,12,13],"blockquote",{},[14,15,16,17,25,26,25,30,33,34],"p",{},"From plugin ",[18,19,21],"a",{"href":20},"\u002Fplugins\u002Faio-claude-toolkit",[22,23,24],"strong",{},"aio-claude-toolkit"," · ",[27,28,29],"code",{},"v2.6.2",[22,31,32],{},"Install:"," ",[27,35,36],{},"\u002Fplugin install aio-claude-toolkit@aiocean-plugins",[38,39,41],"h1",{"id":40},"skillify-capture-session-workflows-as-reusable-skills","Skillify - Capture Session Workflows as Reusable Skills",[14,43,44],{},"Turn what you just did into a skill that works every time.",[46,47,49],"h2",{"id":48},"goal","Goal",[14,51,52],{},"Analyze the current session, interview the user about their intent, and produce a well-structured SKILL.md file that captures the repeatable process.",[46,54,56],{"id":55},"steps","Steps",[58,59,61],"h3",{"id":60},"_1-load-session-memory","1. Load Session Memory",[14,63,64,65,68,69,72],{},"Session memory is stored at ",[27,66,67],{},"~\u002F.claude\u002Fprojects\u002F{sanitized-cwd}\u002F{sessionId}\u002Fsession-memory\u002Fsummary.md"," where sanitized-cwd replaces all non-alphanumeric characters with ",[27,70,71],{},"-",".",[14,74,75],{},"Use Glob to find and Read the most recent session memory for the current project. This gives you a compact summary of the entire session including compacted context. If no session memory exists, skip this step and rely on the current conversation context.",[58,77,79],{"id":78},"_2-analyze-the-session","2. Analyze the Session",[14,81,82],{},"Combine the session memory with the current conversation to identify:",[84,85,86,90,93,96,99,102,105],"ul",{},[87,88,89],"li",{},"What repeatable process was performed",[87,91,92],{},"What the inputs\u002Fparameters were",[87,94,95],{},"The distinct steps (in order)",[87,97,98],{},"The success artifacts\u002Fcriteria for each step (e.g. not just \"writing code\" but \"an open PR with CI fully passing\")",[87,100,101],{},"Where the user corrected or steered you - these are gold for capturing preferences",[87,103,104],{},"What tools and permissions were needed",[87,106,107],{},"What agents were used (if any)",[58,109,111],{"id":110},"_3-interview-the-user","3. Interview the User",[14,113,114],{},"Use AskUserQuestion for ALL questions. Never ask via plain text. For each round, iterate until the user is happy. Always include a freeform \"Other\" option - do NOT add your own \"Needs tweaking\" option.",[14,116,117],{},[22,118,119],{},"Round 1: High-level confirmation",[84,121,122,125],{},[87,123,124],{},"Suggest a name and description for the skill based on your analysis",[87,126,127],{},"Suggest high-level goal(s) and specific success criteria",[14,129,130],{},[22,131,132],{},"Round 2: Details",[84,134,135,138,141,152],{},[87,136,137],{},"Present the high-level steps as a numbered list. Tell the user you will dig into details next round.",[87,139,140],{},"If the skill needs arguments, suggest them based on what you observed.",[87,142,143,144,147,148,151],{},"Ask if this skill should run ",[27,145,146],{},"inline"," (in current conversation, user can steer mid-process) or ",[27,149,150],{},"fork"," (sub-agent with own context, better for self-contained tasks).",[87,153,154,155],{},"Ask where to save. Suggest a default based on context:\n",[84,156,157,167,176],{},[87,158,159,162,163,166],{},[22,160,161],{},"This repo"," (",[27,164,165],{},".claude\u002Fskills\u002F\u003Cname>\u002FSKILL.md",") - for workflows specific to this project",[87,168,169,162,172,175],{},[22,170,171],{},"Personal",[27,173,174],{},"~\u002F.claude\u002Fskills\u002F\u003Cname>\u002FSKILL.md",") - follows you across all repos",[87,177,178,162,181,184],{},[22,179,180],{},"Plugin",[27,182,183],{},"\u003Cplugin-path>\u002Fskills\u002F\u003Cname>\u002FSKILL.md",") - if part of a plugin",[14,186,187,190],{},[22,188,189],{},"Round 3: Breaking down each step","\nFor each major step, if not glaringly obvious, ask:",[84,192,193,196,199,202,205,208],{},[87,194,195],{},"What does this step produce that later steps need? (data, artifacts, IDs)",[87,197,198],{},"What proves this step succeeded, and that we can move on?",[87,200,201],{},"Should the user confirm before proceeding? (especially irreversible actions: merging, sending messages, destructive ops)",[87,203,204],{},"Are any steps independent and could run in parallel?",[87,206,207],{},"How should the skill execute? (direct, Task agent, agent team for concurrent steps)",[87,209,210],{},"What are hard constraints or preferences? Things that must or must not happen?",[14,212,213],{},"Do multiple rounds here if needed - one per step for complex workflows. Iterate as much as needed.",[14,215,216,219],{},[22,217,218],{},"IMPORTANT",": Pay special attention to places where the user corrected you during the session.",[14,221,222],{},[22,223,224],{},"Round 4: Final questions",[84,226,227,230],{},[87,228,229],{},"Confirm when this skill should be invoked, and suggest\u002Fconfirm trigger phrases (e.g. \"cherry-pick to release\", \"CP this PR\", \"hotfix\")",[87,231,232],{},"Ask for any gotchas or things to watch out for",[14,234,235],{},"Stop interviewing once you have enough information. Don't over-ask for simple processes!",[58,237,239],{"id":238},"_4-write-the-skillmd","4. Write the SKILL.md",[14,241,242],{},"Use this format:",[244,245,251],"pre",{"className":246,"code":248,"language":249,"meta":250},[247],"language-markdown","---\nname: {{skill-name}}\ndescription: {{one-line description including trigger phrases}}\nallowed-tools:\n  {{list of tool permission patterns observed during session}}\nwhen_to_use: {{detailed description of when to auto-invoke, starting with \"Use when...\"}}\nargument-hint: \"{{hint showing argument placeholders}}\"\narguments:\n  {{list of argument names}}\ncontext: {{inline or fork -- omit for inline}}\n---\n\n# {{Skill Title}}\n\nDescription of skill\n\n## Inputs\n- `$arg_name`: Description of this input\n\n## Goal\nClearly stated goal for this workflow. Best if you have clearly defined artifacts or criteria for completion.\n\n## Steps\n\n### 1. Step Name\nWhat to do in this step. Be specific and actionable. Include commands when appropriate.\n\n**Success criteria**: ALWAYS include this! Shows the step is done and we can move on.\n\n...\n","markdown","",[27,252,248],{"__ignoreMap":250},[14,254,255,258],{},[22,256,257],{},"Per-step annotations"," (use where helpful):",[84,260,261,267,288,294,300],{},[87,262,263,266],{},[22,264,265],{},"Success criteria"," - REQUIRED on every step",[87,268,269,272,273,276,277,280,281,284,285],{},[22,270,271],{},"Execution",": ",[27,274,275],{},"Direct"," (default), ",[27,278,279],{},"Task agent",", ",[27,282,283],{},"Teammate"," (parallel + inter-agent), or ",[27,286,287],{},"[human]",[87,289,290,293],{},[22,291,292],{},"Artifacts",": Data this step produces that later steps need",[87,295,296,299],{},[22,297,298],{},"Human checkpoint",": When to pause for user confirmation (irreversible actions, error judgment, output review)",[87,301,302,305],{},[22,303,304],{},"Rules",": Hard rules for the workflow. User corrections during the session are especially useful here.",[14,307,308],{},[22,309,310],{},"Step structure tips:",[84,312,313,316,322],{},[87,314,315],{},"Steps that can run concurrently use sub-numbers: 3a, 3b",[87,317,318,319,321],{},"Steps requiring the user to act get ",[27,320,287],{}," in the title",[87,323,324],{},"Keep simple skills simple -- a 2-step skill doesn't need annotations on every step",[14,326,327],{},[22,328,329],{},"Frontmatter rules:",[84,331,332,345,354,360],{},[87,333,334,337,338,341,342],{},[27,335,336],{},"allowed-tools",": Minimum permissions needed. Use patterns like ",[27,339,340],{},"Bash(gh:*)"," not ",[27,343,344],{},"Bash",[87,346,347,350,351,353],{},[27,348,349],{},"context",": Only set ",[27,352,150],{}," for self-contained skills that don't need mid-process user input",[87,355,356,359],{},[27,357,358],{},"when_to_use"," is CRITICAL - tells the model when to auto-invoke. Start with \"Use when...\" and include trigger phrases",[87,361,362,365,366,369,370,373],{},[27,363,364],{},"arguments"," and ",[27,367,368],{},"argument-hint",": Only include if the skill takes parameters. Use ",[27,371,372],{},"$name"," in the body for substitution",[58,375,377],{"id":376},"_5-review-and-save","5. Review and Save",[14,379,380],{},"Before writing the file:",[382,383,384,387,390],"ol",{},[87,385,386],{},"Output the complete SKILL.md content as a yaml code block so the user can review it with syntax highlighting",[87,388,389],{},"Ask for confirmation using AskUserQuestion: \"Does this SKILL.md look good to save?\"",[87,391,392],{},"Only write the file after user confirms",[14,394,395],{},"After writing, tell the user:",[84,397,398,401,407],{},[87,399,400],{},"Where the skill was saved",[87,402,403,404],{},"How to invoke it: ",[27,405,406],{},"\u002F{{skill-name}} [arguments]",[87,408,409],{},"That they can edit the SKILL.md directly to refine it",{"title":250,"searchDepth":411,"depth":411,"links":412},2,[413,414],{"id":48,"depth":411,"text":49},{"id":55,"depth":411,"text":56,"children":415},[416,418,419,420,421],{"id":60,"depth":417,"text":61},3,{"id":78,"depth":417,"text":79},{"id":110,"depth":417,"text":111},{"id":238,"depth":417,"text":239},{"id":376,"depth":417,"text":377},"Capture this session's repeatable process into a reusable SKILL.md file via guided interview.","skill","md",{},true,"\u002Fplugins\u002Faio-claude-toolkit\u002Faio-skillify",{"title":5,"description":422},"plugins\u002Faio-claude-toolkit\u002Faio-skillify","OCwtyHULSRksUxeakvEdK8mNMMEMDUPGo2MpdwG_ZWo",[],1779712313052]