light-pop-portrait
Light flat-vector single-character portrait — a hand-drawn child/tween on a fully saturated flat color background, with chunky black hair, big round eyes, a tiny red nose, and pink five-point-star sparkle blush. Trigger when the user says /light-pop-portrait, asks for a "light pop portrait", "flat-vector character portrait", "saturated color-block character", or briefs with expression + hairstyle + palette + prop.
What this skill does
# /light-pop-portrait — locked light hand-drawn character portrait style
This style produces single-character portrait illustrations on a fully saturated solid color block. The *mood* is light and playful (chunky hair, big shiny eyes, star-sparkle blush, one expressive prop). The *format* is pop (one saturated solid color filling the canvas, ~5-color palette, no gradients). The line quality is **light** — thin, slightly wavy, hand-drawn — never thick or stamped.
Each invocation swaps a small set of dials. The locked frame (character anatomy, line quality, surface, background-as-block, star-sparkle blush) never moves.
## Prompt interpretation
The user will usually give a short brief — sometimes just an expression ("sleepy"), sometimes a pose plus a palette ("cheering arms up, magenta"), sometimes a near-complete mini-portrait ("zen androgynous kid with a beret holding a tiny bird, deep teal"). Translate that into a full portrait brief without stopping to ask:
1. **Pick an expression** — the emotional anchor of the piece (cheerful, sleepy, surprised, sly, laughing, zen, pouty, daydreaming, teary, embarrassed).
2. **Pick a hairstyle** from the catalogue (blunt bob + bangs, twin braids, fluffy afro curls, mushroom bowl cut, spiky bedhead, side-parted slick, high ponytail, double buns, side braid).
3. **Decide gender register** — girl, boy, or androgynous; always in the child-to-tween age range.
4. **Pick a pose** that fits the expression (head-and-shoulders, chest-up, leaning on hand, both hands up, hugging-prop, full-body seated, side profile).
5. **Pick one saturated background hue** — the single color filling the entire canvas behind the character.
6. **Optionally add headwear** (bucket hat, beret, hairband + bow, headband, glasses, none).
7. **Pick one signature prop** that the character interacts with (umbrella, popsicle, cherry, frog, bunny, bird, balloon, lantern, book, instrument).
8. **Pick the outfit register** (striped tee, overalls, sweater, pajamas, raincoat, sundress, hoodie).
9. **Honor the locked frame** — anatomy, line quality, flat fills, saturated background, star-sparkle blush — those never move.
Bias toward playful character moments where one expression + one prop carry the whole composition. Avoid crowded scenes, multiple characters, or heavy environmental staging — this style is a portrait, not a scene.
## Locked style axes (NEVER vary)
### Character anatomy
- **Single character per piece** — never a group, never a duo
- **Child-to-tween proportions** — large head relative to body, shoulders and chest fit comfortably in the frame
- **Round face shape** with a soft pointed chin; cheeks rounded but not exaggerated
- **Big round shiny eyes** — black filled pupils with a single small white glint per eye, thin black lashes above (never below)
- **Tiny round red nose** — a small red circle dot, centered between eyes and mouth, ~1/6 the size of an eye
- **Small expressive mouth** — a short curved line for closed smiles, a small rounded shape for open mouths; teeth shown only for big laughing grins
- **Hair is glossy chunky black** — solid black fill, never gradient, slight glossy highlight allowed as a thin lighter stroke on the crown
- **Hair always has a small flyaway strand or two** — one or two stray wisps sticking out from the silhouette to keep it from looking stamped
### Line quality
- **Thin delicate slightly-wavy black contour lines** — light hand-drawn feel, NOT thick, NOT bold, NOT uniformly stamped
- Lines have small natural irregularities — the contour breathes; perfect arcs feel wrong here
- No double-stroke, but minor variation in line weight along a single contour is welcome
- The whole canvas should read as if drawn by a confident but light hand
### Star-sparkle blush
- **Two pink five-point-star sparkles on the cheeks** — small five-point stars, ONE per cheek, sitting just under the eye
- **NEVER solid round dots** — this is the most commonly-failed locked detail; specify "five-point stars, not circles" if a model drifts
- Stars are a soft warm pink, slightly transparent or fully filled — never red, never saturated magenta
### Surface
- **Flat solid color fills only** — no gradients, no airbrush, no shading, no halftone, no riso grain, no paper texture
- The canvas is plastic-clean — no organic texture anywhere
- Subtle cel-shading is acceptable as a single flat shadow tone on the prop or under the hat brim, but is **optional**; most pieces are pure flat
### Background
- **One fully saturated single hue filling the entire canvas** — no gradient, no vignette, no second color band, no horizon strip, no border, no frame
- The character sits cleanly on top of the color block with no environment behind them
- Hue is one from the palette dial below
### Composition
- Square 1:1 canvas (1024×1024 recommended)
- Single portrait — the character is the only subject; the prop is held by or near the character, not staged independently
- Character occupies the center of the frame, head + shoulders to upper torso minimum, full body for `full-body seated`
- Clean negative space around the silhouette — the color block must breathe
- No text, no logos, no captions, no headlines
### Palette discipline
- **Limit ~5 total colors per piece** — background + hair-black + skin tone + 1-2 outfit colors + the prop's primary color
- Skin is a soft warm beige or peach, slightly desaturated; never bright orange, never pale white
- Hair is always solid black with optional one-stroke gloss
## Variable axes (the eight dials)
These are the only things that should change between pieces.
| # | Axis | What it controls | Example values |
|---|---|---|---|
| 1 | **Expression** | The emotional anchor | cheerful · sleepy/yawning · surprised "O" · sly wink · laughing (open grin) · zen closed-eye smile · pouty · daydreaming · teary · embarrassed |
| 2 | **Hairstyle** | Hair silhouette + cut | blunt bob + bangs · twin braids · fluffy afro curls · mushroom bowl cut · spiky bedhead · side-parted slick · high ponytail · double buns · side braid with bow |
| 3 | **Gender / Identity** | Gender register | girl · boy · androgynous (all in child–tween age range) |
| 4 | **Pose** | Body language | head-and-shoulders portrait · chest-up · leaning on hand · both hands up · hugging the prop · full-body seated · side profile |
| 5 | **Color palette** | The single saturated background hue | cobalt blue · mint green · peach coral · lavender purple · hot magenta · deep forest teal · mustard yellow · terracotta · dusty rose · navy |
| 6 | **Headwear** | What sits on the head (optional) | bucket hat · beret · hairband + bow · headband · round glasses · none |
| 7 | **Prop** | The one object the character interacts with | small umbrella · popsicle · cherry · frog · bunny · songbird · balloon · paper lantern · book · tiny instrument |
| 8 | **Outfit register** | Clothing baseline | striped tee · denim overalls · cozy sweater · pajamas · raincoat · sundress · hoodie |
## Brief template
When generating, expand the user's input into this internal brief before describing the image to the model:
```
Expression: <one phrase — what emotion the face carries>
Hairstyle: <cut + length + any small detail like a braid or bow>
Gender register: <girl / boy / androgynous, child–tween>
Pose: <verb + body language>
Background hue: <one saturated color filling the canvas>
Headwear: <bucket hat / beret / hairband / glasses / none>
Prop: <the one object held or near the character>
Outfit: <one or two layered pieces in a flat palette>
Color count: 5 max
```
## Worked examples
Six locked-style portraits, each holding the frame and varying the eight dials.
### Example 1 — cheerful girl · yellow bucket hat · mushroom umbrella · cobalt

- Expression: cheerful, small upturned smile, big shiny eyes
- Hairstyle: blunt bob with bangs + one tiny side braid tied with a pink bow
- Gender: girl
- PoRelated in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.