Spec Kit Agent Skills Model Card
这是可索引项目报告证据页：它保留 Spec Kit Agent Skills Model Card 的源材料入口、机制线索和限制提醒；正文仍需 reader/editor 与 academic public-copy review 后才能当作最终结论引用。
Spec Kit Agent Skills Model Card
| Field | Value |
|---|---|
| Repository | dceoy/speckit-agent-skills |
| Category | Spec-Driven Agent Workflow Skills |
| Stars / forks snapshot | 88 / 21 |
| Language | Shell |
| License | AGPL-3.0 |
| Raw capture | raw-github/dceoy_speckit-agent-skills.md |
| Updated by | hourly public metadata update, 2026-05-26 07:45 +0800 |
1. Role in Self Evolve
speckit-agent-skills maps Spec Kit methodology into shared skills, prompts, commands and agent files for Claude Code, Codex CLI, GitHub Copilot CLI, Gemini CLI and related runtimes. It matters to self-evolving agents because durable improvement often appears first as reusable procedures, workflow discipline and domain-specific skill assets rather than direct model-weight updates.
2. Working Principle
spec-driven workflow -> constitution/specify/plan/tasks/implement skills -> multi-runtime agent process discipline
3. Evidence Path
web-observed GitHub page showed 88 stars, 21 forks, 92 commits, AGPL-3.0 license, Shell primary language, v0.1.2 release on 2026-03-14, and README sections for shared speckit-* skills plus runtime-specific Claude Code, Codex, GitHub Copilot CLI and Gemini CLI access. Shell GitHub API access remained blocked by DNS and local gh auth was invalid, so this card treats the snapshot as web-observed rather than API-verified.
4. Teaching Use
Use this card to explain Spec-Driven Agent Workflow Skills: the project helps show how skill libraries turn repeated expert behavior into installable agent assets that can be versioned, reviewed and reused across agents.
5. Limits
The repository was not cloned in this iteration; no benchmark, plugin install, workflow run, trading analysis, Spec Kit task, or agent loop was executed. Counts and claims are visible public-page/search signals unless independently revalidated later.