readable-verilog-generator
VerifiedUse when creating, writing, reviewing, annotating, repairing, refactoring, or validating readable Verilog RTL, including synthesizable Verilog-2001 .v files, existing-RTL analysis, semantic comment annotation, testbench scaffold planning, ASIC-quality review, local or remote Vivado/xsim validation, evidence-backed repair, and workflow trace diagnosis.
references/skill/skill-standards.md
View on GitHub# Skill Development Standards
This reference captures the active skill-authoring standards used by this repository.
## Frontmatter Rules
- Keep exactly two frontmatter fields in `SKILL.md`: `name` and `description`.
- Keep frontmatter under 1024 characters total.
- Keep `name` lowercase with letters, numbers, and hyphens only.
- Start `description` with `Use when`.
- Keep `description` focused on trigger conditions, symptoms, and contexts.
- Do not summarize the internal workflow in `description`.
- Keep `description` at or below 500 characters.
## Progressive Disclosure Rules
- Keep the main `SKILL.md` focused on trigger conditions, workflow boundaries, and resource navigation.
- Move heavy reference material into `references/`.
- Move reusable scripts, templates, and fixtures into their own folders instead of duplicating large inline content.
- Every `Load ... when ...` resource path named in `SKILL.md` must exist inside the skill directory.
## Pattern Rules
This skill should explicitly preserve the following design-pattern mix:
- Tool Wrapper
- Generator
- Reviewer
- Inversion
- Pipeline
## Evaluation Rules
- Verification claims must be backed by fresh command output.
- Skill-level checks must cover metadata, resource integrity, workflow behavior, and confidence gates.
- Generated Verilog validation must include a hard same-line semantic comment-placement gate for every non-empty RTL and testbench `.v` code line, with construct-level coverage metrics and negative fixtures for generic or shared-adjacent comments.
- Effectiveness evidence should compare behavior with and without the skill when a realistic evaluation harness exists.
- Pass-rate delta and failure-mode coverage are preferred over anecdotal success claims.
## Packaging Rules
- Markdown inside the installable skill must remain ASCII-only by default; only exact paths listed in `validation.markdown_non_ascii_allowlist` may contain non-ASCII text.
- Installable content must not keep hard dependencies on temporary external reference directories.
- Public skill interfaces must stay stable when internal runtime details evolve.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| planning-and-task-breakdownaddyosmani/agent-skillsBreaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task feels too large to start, when you need to estimate scope, or when parallel work is possible. | Stars: 119k | Updated | Claude CodeCodex | Verified | Install |
| screenpipethedotmack/claude-memSearch screen and audio history recorded by Screenpipe and optionally save selected context into claude-mem. Use when the user asks to recall desktop activity, a meeting, or something seen in another app through Screenpipe. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |
| contact-widgetnexu-io/open-designSelf-contained floating chat widget with welcome screen, social links, meeting button, and message input. Single HTML file, zero dependencies. | Stars: 108k | Updated | Claude CodeCodex | Verified | Install |
| meeting-notesnexu-io/open-designMeeting notes page — title bar with attendees, agenda checklist, decisions block, action items table with owners + dates, and a "next meeting" footer. Use when the brief mentions "meeting notes", "minutes", "1:1 notes", "all-hands recap", or "会议纪要". | Stars: 108k | Updated | Claude CodeCodex | Verified | Install |