跳到正文
cccskills
免费在 GitHub 公开

bootstrap-processor-project

Initialize or safely upgrade a processor project's root AGENTS.md from a maintained baseline. Use when starting a processor project, adding project-level Agent collaboration rules, or comparing an existing AGENTS.md with the package baseline. This skill only handles AGENTS.md; it does not scaffold documentation, inspect or configure environments, install tools, or modify processor source.

中文简介正在准备中。当前显示原文说明。

sixblade325/WaterHand-Processor-Development-Skills
查看安装方法

包含的文件（3）

  • SKILL.md5.7 KB
  • agents/openai.yaml270 B
  • assets/AGENTS.md3.0 KB

SKILL.md（原文）

安装前可以查看交给 AI 代理的指令内容。

Bootstrap Processor Project

Create one human-editable, project-owned AGENTS.md at the processor project root. Preserve existing project authority and keep every other project file unchanged.

Baseline

Read assets/AGENTS.md completely before drafting or comparing a project file. Treat it as a maintained output template. It becomes project authority only after the user accepts it or it is written into a project that lacks AGENTS.md under an explicit bootstrap request.

Keep this maintained baseline within 4096 UTF-8 bytes. It owns only project authority, authorization, path mapping, verified tool entrypoints, and concise routing to task Skills. Detailed design gates, hardware rules, verification checklists, and source-summary requirements belong to the corresponding Skills and their references. Do not copy them or personal conversational style rules into the generic baseline. Project-owned files may grow to record evidenced local constraints; the package budget is not a limit on user projects.

On Windows PowerShell, read UTF-8 project documents with Get-Content -Raw -Encoding utf8 -LiteralPath <path>. Never depend on the Windows PowerShell 5.1 default file encoding. Apply this rule before inspecting an existing AGENTS.md and throughout the bootstrap task.

Boundary

  1. This skill may create or revise only the target project's root AGENTS.md.
  2. Do not create Architecture, Design, Source, Verification, runtime, configuration, or placeholder files.
  3. Do not inspect, install, configure, repair, or validate the development environment. Environment and toolchain responsibilities belong to deterministic scripts outside this skill.
  4. Do not run builds, tests, simulators, synthesis tools, package managers, or environment setup commands.
  5. Do not modify nested AGENTS.md files unless the user explicitly names one as the target.
  6. Do not add machine-specific absolute paths, usernames, local tool locations, project-specific processor facts, or inferred architecture decisions.
  7. After creation, the user project owns its AGENTS.md. Never synchronize or overwrite it from a later package version automatically.

Resolve the target

  1. Use the project root explicitly named by the user.
  2. When no path is named, use the current Git worktree root if it is unambiguous and within the user's stated scope.
  3. If multiple repositories or possible roots remain, ask the user to identify the target before writing.
  4. Read all applicable existing AGENTS.md instructions before examining or changing the target.

Missing AGENTS.md

  1. Inspect the repository layout and tracked project documentation read-only.
  2. Copy the baseline to <project-root>/AGENTS.md.
  3. Adapt only facts directly supported by the repository or an explicit user decision:
    • project title;
    • Architecture, Design, Source, Verification, and Runtime path mapping;
    • declared build, test, simulation, synthesis, and static-check script entrypoints.
  4. Retain the baseline authority, authorization, evidence, and project-ownership boundaries. Add only evidenced local constraints within the bootstrap request.
  5. Use the baseline mapping when no contrary project evidence exists: documentation under doc/, source under src/, and runtime data under .runtime/. This matches the new-project default of organize-processor-docs; creating this file does not create those directories. Preserve an existing project's documented mapping and adapt task references to it. Do not invent commands or paths.
  6. Write no other file.

An explicit request to initialize or bootstrap the named project authorizes creating a missing root AGENTS.md. It does not authorize any other project or environment change.

Existing AGENTS.md

  1. Read the existing file and treat its current project rules as authoritative.
  2. Do not replace it with the baseline.
  3. Compare by responsibility: project authority, authorization, path mapping, verified tool entrypoints, local constraints, and task-Skill routing. Identify copied methodology as a simplification proposal, keeping the existing project file unchanged until authorized.
  4. Preserve every existing rule unless the user explicitly approves its removal or replacement.
  5. Report proposed additions, conflicts, and obsolete project-specific rules as an incremental change set.
  6. Apply the change set only after the user explicitly authorizes the revision.

Acceptance

Before completing, verify:

  1. At most one target file changed, <project-root>/AGENTS.md.
  2. A pre-existing file was not overwritten or reduced without explicit authorization.
  3. The file contains no unresolved scaffold markers, guessed commands, or machine-specific paths.
  4. The authority rules distinguish explicit user decisions, current Git authorities, implementation evidence, and external references.
  5. The file retains authorization, evidence, and no-auto-overwrite constraints. New-project mappings compose with the documentation Skill; existing mappings remain intact unless migration was authorized.
  6. No environment command or unrelated project action ran.

Report the target path, whether the operation created or revised the file, the evidence-backed adaptations, validation performed, and unresolved user choices.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问，或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护，适用 Google 的隐私权政策和服务条款。

同一仓库的技能

design-chisel-processor

免费

Develop, challenge, and document Chisel processor microarchitecture designs before implementation. Use when discussing or writing design documents for processor pipelines, queues, issue logic, rename, ROB, LSU, caches, MSHRs, forwarding, wakeup, replay, flush, privilege, exceptions, or other cycle-accurate hardware mechanisms; when converting design conversations into stable Markdown specifications; or when reviewing a proposed Chisel CPU design for correctness, timing, area, verification cost, and cross-document consistency.

中文简介正在准备中。当前显示原文说明。

sixblade325/WaterHand-Processor-Development-Skills152026年10月3日 更新

implement-chisel-processor

免费

Document-driven workflow for implementing, reviewing, and verifying Chisel processor and memory-subsystem RTL. Use when Codex must work from maintained Architecture and Design, trace a complete integration surface, reason in synthesized-hardware terms, keep source-adjacent _codex.md summaries current, avoid redundant or overprotective logic, or run focused functional verification.

中文简介正在准备中。当前显示原文说明。

sixblade325/WaterHand-Processor-Development-Skills152026年10月3日 更新

optimize-chisel-fpga-timing

免费

Diagnose and optimize timing-critical Chisel RTL for FPGA implementation while preserving cycle semantics. Use for Vivado timing bottlenecks, long ready or admission paths, queue and issue selection, free-list bank mapping, priority encoders, one-hot arbitration, wide muxes, late-arriving forwarding or override data, high-fanout controls, cross-module predicates, register-boundary changes, emitted-Verilog inspection, or routed-DCP A/B analysis. Also use when a source-level simplification needs proof that it changes timing without changing architectural state.

中文简介正在准备中。当前显示原文说明。

sixblade325/WaterHand-Processor-Development-Skills152026年10月3日 更新

organize-processor-docs

免费

Establish, author, restructure, or audit human-first processor Architecture, Design, Research, Review, and Verification documentation. Use for progressive documentation scaffolding, authority maps, reading paths, document-type content contracts, length-budgeted splitting, or maintainability reviews. Do not use as a substitute for cycle-accurate microarchitecture analysis or RTL implementation.

中文简介正在准备中。当前显示原文说明。

sixblade325/WaterHand-Processor-Development-Skills152026年10月3日 更新

trace-vivado-timing-to-rtl

免费

Task-sized forensic analysis of Vivado synthesis and routed timing evidence for processor RTL. Use when Codex must trace a named setup or hold path, audit whole-design timing populations, compare implementation runs, map primitive, LUT, CARRY, BRAM, DSP, MUXF, register, and routed-net stages back to Chisel or generated RTL, distinguish logic from routing pressure, identify missing DCP queries, or write an evidence-backed timing report. This skill is read-only by default and stops at ranked modification directions. Use optimize-chisel-fpga-timing for RTL edits and routed A/B implementation closure.

中文简介正在准备中。当前显示原文说明。

sixblade325/WaterHand-Processor-Development-Skills152026年10月3日 更新

查看 sixblade325 的全部技能

报告此技能的问题