Bỏ qua tới nội dung
TakoAPI
Quay lại kỹ năng

skeall

bởi dorukardahan

Coding Agents & IDEs
429Lượt tải do clawskills.sh công bố
1Lượt xem
0 lượt thíchĐăng nhập để thích kỹ năng này
—Chưa có đánh giá
Đánh giá của bạn

Tóm tắt

Agent Skills (SKILL.md) builder, auditor, and improver for cross-platform LLM agents.

Mô tả

Kỹ năng này làm gì

Skeall creates, audits, and improves Agent Skills (SKILL.md files) following the agentskills.io open standard. It interviews users to scaffold new skills, runs compliance scans with severity-tagged reports, and refactors existing skills for cross-platform LLM compatibility.

It encodes the full agentskills.io spec and real-world cross-platform testing lessons into one auditable workflow, replacing manual checklist review.

Khi nào nên dùng

  • Scaffolding a new SKILL.md via guided interview
  • Auditing an existing skill for spec violations before publishing
  • Batch scanning all skills in a directory for quality issues
  • Fixing persona-based framing or token bloat in an inherited skill
  • Verifying cross-platform compatibility before sharing a skill publicly

Quy trình ví dụ

ĐẦU VÀO

User asks: /skeall --scan ./my-skill

AGENT
  1. 1

    Read SKILL.md and list directory structure at the target path

  2. 2

    Run all checklist checks across structure, frontmatter, content, LLM-friendliness, security, and cross-platform categories

  3. 3

    Tag each check as PASS, WARN, or FAIL with HIGH/MEDIUM/LOW severity

  4. 4

    Calculate score using the formula: 10 - (HIGHs x 1.5) - (MEDIUMs x 0.5) - (LOWs x 0.2)

  5. 5

    Output severity-tagged report grouped by category with a suggested next command

ĐẦU RA

Audit report showing score (e.g. 6.5/10), all failing checks grouped by severity, and a prompt to run --improve if HIGH issues are found

Cài đặt

clawhub install dorukardahan/skeall

Chạy lệnh CLI, hoặc dán prompt vào Claude Code, Cursor, Codex hay coding agent bạn đang dùng.

Liên kết