agentsclimarketplace

Skills creator

Skill quynj/quynj-claw/.agents/skills/skills-creator

本地 Agent 会话控制台 — 基于 AgentScope Java 的轻量级 AI 对话运行时

Install
npx -y skills add quynj/quynj-claw --skill skills-creator

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Create or update project-local Quynj Claw skills under .agents/skills. Use when the user asks to add a new skill, revise an existing SKILL.md, define trigger descriptions, add agents/openai.yaml metadata, or turn stable system-prompt guidance into progressive skill instructions.

SKILL.md

2.7 KB, as published. Nobody here has run it

Skills Creator

Purpose

Create concise, reliable project-local skills that AgentScope can load from .agents/skills. Prefer skills for reusable procedures, durable operating guides, and specialized workflows that do not need to live in the default system prompt.

Skill Shape

Every skill must include:

  • SKILL.md with YAML frontmatter containing only name and description.
  • A folder name exactly matching the skill name.
  • A lowercase hyphenated skill name.

Recommended when useful:

  • agents/openai.yaml for UI-facing metadata.
  • scripts/, references/, or assets/ only when they directly support the skill.

Do not create extra README, changelog, installation notes, or unrelated docs inside a skill.

Workflow

  1. Clarify the skill's trigger and expected tasks from the user request.
  2. Choose a short hyphenated name. If the user gives a name, preserve it unless invalid.
  3. Create the skill under .agents/skills/<skill-name>/.
  4. Write a compact description that includes both what the skill does and when to use it, because this is the trigger text.
  5. Keep SKILL.md body procedural and short. Put only information another agent needs at execution time.
  6. Add agents/openai.yaml with:
    • display_name
    • short_description between 25 and 64 characters
    • default_prompt that explicitly mentions $<skill-name>
  7. Validate the skill structure. If the validation script is unavailable, at least check that frontmatter exists and has name and description.
  8. Update project docs or the default system prompt if the new skill changes the advertised project-local skills.

Quynj Claw Constraints

  • Use only file tools registered in this runtime: view_text_file, write_text_file, insert_text_file, list_files, and list_directory.
  • Do not reference unavailable tools such as edit_file unless they are explicitly added to the runtime.
  • Keep skills project-local unless the user asks for a global Codex skill.
  • Do not put runtime state in .agents/skills; use .agents/memory for durable memory and .agents/ui-store only for UI projections.

Good Skill Descriptions

Use descriptions like:

Create conventional commit messages and commits for this repository. Use when the user asks to commit, stage, amend, or prepare a PR commit summary.

Avoid descriptions like:

Helpful notes for commits.

The first form gives the loader clear trigger conditions; the second does not.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.