agentleFS
Sign inStart free
QuixiAIQuixiAI/agi-memory/skills/installed/image-gen/SKILL.md

image-gen

Skill617 starsChanged 10 months agoMITView on GitHub

Generate images from text descriptions using DALL-E or compatible providers

  • Reads credentials

We split this skill into the rules it gives an agent

12 rules. 0 are common, 11 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 27 edited versions we read keep these.

  • 11

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create images from natural language descriptions using AI image generation models (DALL-E 3 or compatible).

    line 15
  2. Distinctive:

    When the user explicitly requests an image ("draw me", "generate a picture of", "create an image")

    line 19
  3. Distinctive:

    When a creative goal involves visual output (social media graphics, concept art, illustrations)

    line 20
  4. Distinctive:

    When a meeting or presentation needs a quick visual and the user asks for one

    line 21
  5. Distinctive:

    Never generate images speculatively during heartbeats unless a goal explicitly calls for it

    line 22
  6. Distinctive:

    Clarify the request: If the user's description is vague, ask a follow-up before generating.

    line 26
  7. Very specific:

    Generate: Call generate_image with the crafted prompt.

    line 34
  8. Distinctive:

    Store if valuable: If the generated image is tied to a goal or project, store a memory noting what was generated and the prompt used, so it …

    line 36
  9. Distinctive:

    Always respect content policies.

    line 40
  10. Distinctive:

    Do not generate images during heartbeats unless an active goal specifically requires visual output.

    line 42
  11. Distinctive:

    When the user asks for a specific style, honor it precisely rather than defaulting to photorealistic.

    line 43
  12. Distinctive:

    If generation fails (API error, content policy rejection), explain the failure clearly and suggest prompt modifications.

    line 44

“Common” is measured: kept by at least 6 in 10 of the 27 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Image Generation Workflow

Create images from natural language descriptions using AI image generation models (DALL-E 3 or compatible).

When to Use

  • When the user explicitly requests an image ("draw me", "generate a picture of", "create an image")
  • When a creative goal involves visual output (social media graphics, concept art, illustrations)
  • When a meeting or presentation needs a quick visual and the user asks for one
  • Never generate images speculatively during heartbeats unless a goal explicitly calls for it

Step-by-Step Methodology

  1. Clarify the request: If the user's description is vague, ask a follow-up before generating. A clear prompt produces vastly better results than a guess. Key dimensions to clarify: subject, style, mood, aspect ratio, and any text that should appear.
  2. Craft the prompt: Translate the user's intent into a detailed image generation prompt. Good prompts include:
    • The primary subject and its pose or arrangement
    • Art style or medium (photorealistic, watercolor, vector illustration, etc.)
    • Lighting and color palette
    • Composition notes (close-up, wide shot, centered, rule of thirds)
    • What to exclude (negative prompt elements phrased positively, e.g., "clean background" instead of "no clutter")
  3. Set parameters: Choose the appropriate size and quality setting. Default to standard quality unless the user requests high quality. Respect the cost implications -- high-quality, large images are more expensive.
  4. Generate: Call generate_image with the crafted prompt. The tool returns a URL or local path to the generated image.
  5. Present and iterate: Show the result to the user. If they want adjustments, refine the prompt and regenerate. Keep the previous prompt as a starting point to maintain continuity.
  6. Store if valuable: If the generated image is tied to a goal or project, store a memory noting what was generated and the prompt used, so it can be reproduced or iterated on later.

Quality Guidelines

  • Always respect content policies. Do not generate images of real people, violent content, or anything the user has not explicitly requested.
  • Be transparent about cost. Image generation consumes API credits; inform the user if they are making many requests.
  • Do not generate images during heartbeats unless an active goal specifically requires visual output. Image generation is expensive.
  • When the user asks for a specific style, honor it precisely rather than defaulting to photorealistic.
  • If generation fails (API error, content policy rejection), explain the failure clearly and suggest prompt modifications.
  • Store the successful prompt alongside the image reference so the user can reproduce or refine the result later.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free