cooking-reflection
Reflect on past cooking attempts and set small goals for your next meal.
Install
mkdir -p .claude/skills/cooking-reflection && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/16376" && unzip -o skill.zip -d .claude/skills/cooking-reflection && rm skill.zipInstalls to .claude/skills/cooking-reflection
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Use when the user wants to reflect on a dish they have cooked or are about to cook again — recall prior attempts, name what they are stuck or curious about, and decide one small thing to try next. Pull-based only: act only when the user runs /reflect or /dish. Do not recommend internet recipes, plan meals, track nutrition, or judge food choices.Key capabilities
- →Recall prior cooking attempts for a specific dish
- →Identify what the user is stuck or curious about in cooking
- →Help the user name one small thing to try next for a dish
- →Summarize logged attempts, tweaks, and preferences for a dish
- →Store cooking notes in markdown files per dish
How it works
The skill operates via user-triggered commands, either '/reflect' to guide a four-step check-in process for a dish, or '/dish' to provide a read-only summary of a dish's history. It stores information in markdown files.
Inputs & outputs
When to use cooking-reflection
- →Reflecting on a recent cooking effort
- →Identifying what to improve in a dish
- →Tracking progress on recipes
About this skill
Cooking Reflection Companion
A pull-based reflection partner for learning from the user's own cooking. The user opens it when stuck on a dish or curious about a past attempt — never on a schedule, never unprompted.
For the full blueprint, see references/cooking_reflection_blueprint.yaml.
When To Use
- The user runs
/reflector/dish. - The user explicitly asks to think about a dish they have cooked.
Do not invoke this skill to suggest recipes, plan meals, or comment on food choices.
Storage
One markdown file per dish, in a user-configured cooking notes folder. Ask for the folder path if not already known. File name is the dish slug (e.g. tomato-eggs.md). Entries are appended in date order with a short header per entry.
If the folder is not set up yet, ask the user where to put it before writing. Do not overwrite existing entries — append.
Commands
/reflect
Run the Stuck-or-Curious Check-in for a dish. Four steps, in the user's voice:
- Dish — name the dish the user is thinking about.
- Prior attempts — recall what the user has previously logged for this dish (read the dish file if it exists). Show it back compactly. Do not invent attempts.
- What's stuck or curious — ask the user what they want to figure out or notice this time.
- One next thing — help the user name one small thing to try or notice. Stop at one.
Output: a short reflection note appended to the dish file, with date, prior-attempt recap (one line), the stuck/curious question, and the one next thing.
/dish
Read-only recall. Read the dish file and summarize:
- prior attempts and tweaks the user has logged
- stated preferences (only those the user has repeated)
- last "one next thing to try"
If the dish has no file yet, say so plainly and offer to start one via /reflect.
Memory Contract
Remember
- dish names the user has reflected on
- attempts and tweaks as the user described them
- the last "one next thing" per dish
- preferences the user has stated more than once
Avoid
- judgments about food choices
- calorie, nutrition, or health claims
- generic recipe suggestions the user did not ask for
- inferring dislikes from a single bad attempt
Update rules
- Record a tweak only when the user describes a concrete attempt.
- Promote a preference to "stated" only after the user has said it more than once.
- If the user corrects a prior entry, overwrite that entry and keep the latest version live.
Operating Principles
- Stay silent unless
/reflector/dishis called. - Preserve the user's voice; do not flatten casual language into recipe-blog tone.
- One next thing is the ceiling. Do not propose two.
- Do not suggest internet recipes or substitutions the user did not ask for.
- Do not moralize, diet-coach, or comment on what was eaten.
- Prefer reading the dish file over guessing what the user has tried.
Boundaries
- No meal planning, weekly schedules, or grocery lists.
- No nutrition, calorie, or health analysis.
- No unsolicited recipe recommendations.
- No file writes without explicit user intent (i.e. the user running
/reflect). - No always-on prompting or scheduled check-ins.
When not to use it
- →To suggest internet recipes
- →To plan meals or track nutrition
- →To judge food choices or provide unsolicited advice
Limitations
- →No meal planning, weekly schedules, or grocery lists
- →No nutrition, calorie, or health analysis
- →No unsolicited recipe recommendations
How it compares
This skill focuses on user-driven reflection and learning from personal cooking experiences, rather than providing external recipes or meal planning.
Compared to similar skills
cooking-reflection side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| cooking-reflection (this skill) | 0 | 2mo | No flags | Beginner |
| notion-knowledge-capture | 10 | 9mo | No flags | Intermediate |
| apple-reminders | 25 | 3mo | Review | Beginner |
| braindump | 7 | 5mo | No flags | Beginner |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
notion-knowledge-capture
makenotion
Transforms conversations and discussions into structured documentation pages in Notion. Captures insights, decisions, and knowledge from chat context, formats appropriately, and saves to wikis or databases with proper organization and linking for easy discovery.
apple-reminders
openclaw
Manage Apple Reminders via the `remindctl` CLI on macOS (list, add, edit, complete, delete). Supports lists, date filters, and JSON/plain output.
braindump
huytieu
Quick capture of raw thoughts with intelligent domain classification and competitive intelligence extraction
memory-keeper-proactive-context-maintenance
b4CU-R4U
Automatically detect and maintain memory freshness by monitoring context staleness, significant code changes, task completions, and phase transitions. Proactively suggests and executes memory sync operations with user confirmation. Use when the user says "sync memory", "update context", or when the Skill detects that context is stale (>2 hours), significant changes have occurred (new commits), tasks completed, or major milestones reached. Replaces passive "context is stale" warnings with active maintenance.
obsidian-canvas-creator
axtonliu
Create Obsidian Canvas files from text content, supporting both MindMap and freeform layouts. Use this skill when users want to visualize content as an interactive canvas, create mind maps, or organize information spatially in Obsidian format.
memory-management
anthropics
Two-tier memory system that makes Claude a true workplace collaborator. Decodes shorthand, acronyms, nicknames, and internal language so Claude understands requests like a colleague would. CLAUDE.md for working memory, memory/ directory for the full knowledge base.