Karpathy LLM Wiki
Astro-Han/karpathy-llm-wiki
A skill your agent uses when building or maintaining a personal LLM-powered knowledge base.
当 AI agent 需要在芯片验证任务中复用持续记忆时使用：查询或维护由 XWIKIDIR 指向的验证项目 LLM wiki，了解验证环境、DUT 功能、 接口、testbench、sequence、checker、coverage、workflow、debug 入口、 BT/IT/ST/SoC 项目上下文，并在用户明确要求、项目 AGENTS.md 授权或任务明确…
Install Xwiki
$ npx skills add BLANK2077/xverif --skill xwiki -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install BLANK2077/xverif xwiki --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/BLANK2077/xverif.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/xwiki .claude/skills/xwiki && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "xwiki" agent skill from https://github.com/BLANK2077/xverif/tree/master/skills/xwiki into .claude/skills/xwiki/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "xwiki", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/BLANK2077/xverif/tree/master/skills/xwikiType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add BLANK2077/xverif --skill xwiki -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install BLANK2077/xverif xwiki --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/BLANK2077/xverif.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/xwiki .agents/skills/xwiki && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "xwiki" agent skill from https://github.com/BLANK2077/xverif/tree/master/skills/xwiki into .agents/skills/xwiki/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "xwiki", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add BLANK2077/xverif --skill xwiki -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install BLANK2077/xverif xwiki --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/BLANK2077/xverif.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/xwiki .cursor/skills/xwiki && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "xwiki" agent skill from https://github.com/BLANK2077/xverif/tree/master/skills/xwiki into .cursor/skills/xwiki/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "xwiki", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/BLANK2077/xverif.git --path skills/xwiki--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add BLANK2077/xverif --skill xwiki -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install BLANK2077/xverif xwiki --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/BLANK2077/xverif.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/xwiki .gemini/skills/xwiki && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "xwiki" agent skill from https://github.com/BLANK2077/xverif/tree/master/skills/xwiki into .gemini/skills/xwiki/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "xwiki", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install BLANK2077/xverif xwikiInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add BLANK2077/xverif --skill xwiki -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/BLANK2077/xverif.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/xwiki .github/skills/xwiki && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "xwiki" agent skill from https://github.com/BLANK2077/xverif/tree/master/skills/xwiki into .github/skills/xwiki/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "xwiki", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add BLANK2077/xverif --skill xwiki -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install BLANK2077/xverif xwiki --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/BLANK2077/xverif.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/xwiki .opencode/skills/xwiki && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "xwiki" agent skill from https://github.com/BLANK2077/xverif/tree/master/skills/xwiki into .opencode/skills/xwiki/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "xwiki", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
xwiki- GitHub stars
- 108
- Token cost
- ~1.4k tokens
- SKILL.md length
- 438 words
- Files
- 69 (incl. scripts, references)
- Skills in repo
- 5
- Repo updated
- First seen
- Licence
- MIT
At a glance
当 AI agent 需要在芯片验证任务中复用持续记忆时使用：查询或维护由 XWIKIDIR 指向的验证项目 LLM wiki，了解验证环境、DUT 功能、 接口、testbench、sequence、checker、coverage、workflow、debug 入口、 BT/IT/ST/SoC 项目上下文，并在用户明确要求、项目 AGENTS.md 授权或任务明确…
- Works in 6 steps: 读取环境变量 XWIKI_DIR，它是当前 session 的 xwiki… → 如果 XWIKI_DIR 未定义或为空，必须询问用户提供路径；不要… → 第一次为项目 ingest 持续记忆时，如果用户没有告知 spec 路径或… → …
- Tasks that involve LLM wikis
- SKILL.md covers 必须先做, 何时使用, 查询顺序 and 编译要求, plus 2 more sections
- Calls python
What it does
Xwiki is an agent skill from BLANK2077/xverif. 当 AI agent 需要在芯片验证任务中复用持续记忆时使用：查询或维护由 XWIKIDIR 指向的验证项目 LLM wiki，了解验证环境、DUT 功能、 接口、testbench、sequence、checker、coverage、workflow、debug 入口、 BT/IT/ST/SoC 项目上下文，并在用户明确要求、项目 AGENTS.md 授权或任务明确 包含知识沉淀时把稳定发现编译回 wiki，避免每次 session 从 0 启动。
Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. The skill folder holds 73 other files, including scripts and reference files (for example `agents/openai.yaml`, `references/compile-process.md` and `references/prompt-output-requirements.md`).
It sits in Knowledge Management, covering LLM wikis. The repository describes itself as: xverif is a local toolkit for chip-verification debug agents. It contains deterministic tools for design and waveform debug, coverage, bit calculations, structured entry…. The licence is MIT.
When your agent uses it
- Tasks that involve LLM wikis
Example prompts
- “/xwiki”
Requirements
- Python 3
Workflow steps
6 steps, taken from the first numbered list in SKILL.md.
- 读取环境变量 XWIKI_DIR，它是当前 session 的 xwiki wiki 根目录。
- 如果 XWIKI_DIR 未定义或为空，必须询问用户提供路径；不要 fallback 到 doc/、当前目录或其他猜测路径。
- 第一次为项目 ingest 持续记忆时，如果用户没有告知 spec 路径或 RTL 路径，必须询问用户；不要自己猜测 spec/RTL 根目录。仅创建空 wiki 骨架不要求提供这些路径。
- 需要校验格式时运行
- 查询 wiki 可隐式触发；创建、修改、废弃或写回页面必须有用户明确要求、项目规则授权或任务范围授权。未授权时只在最终结果中列出建议写回的稳定结论。
- 初始化空 wiki 目录时优先运行
What it can do on your machine
Read from SKILL.md and the folder at commit 6aa13f9. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships 1 file in scripts/, which the agent can run.
Shell commands in SKILL.md call:
python
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Xwiki loads about 1.4k tokens when it runs, and up to ~13k if it reads all its reference files. Until then it costs about 59 tokens; SKILL.md has 438 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
SKILL.md
The full file from BLANK2077/xverif at commit 6aa13f9, republished under its MIT licence (© BLANK2077). 438 words, ~1,419 tokens.
Download SKILL.mdSave it as
.claude/skills/xwiki/SKILL.md (or your agent's skills folder). This skill also uses 68 other files; get the full folder from GitHub.- name
- xwiki
- description
- 当 AI agent 需要在芯片验证任务中复用持续记忆时使用：查询或维护由 XWIKI_DIR 指向的验证项目 LLM wiki，了解验证环境、DUT 功能、 接口、testbench、sequence、checker、coverage、workflow、debug 入口、 BT/IT/ST/SoC 项目上下文，并在用户明确要求、项目 AGENTS.md 授权或任务明确 包含知识沉淀时把稳定发现编译回 wiki，避免每次 session 从 0 启动。
xwiki 持续记忆 Skill
xwiki 用来维护芯片验证项目的 LLM wiki。它不是 CLI/MCP 工具，也不读取旧版运行时状态目录；它规定 AI 如何查询、增删改查和校验一个持久 Markdown wiki。
必须先做
- 读取环境变量
XWIKI_DIR，它是当前 session 的 xwiki wiki 根目录。 - 如果
XWIKI_DIR未定义或为空，必须询问用户提供路径；不要 fallback 到doc/、当前目录或其他猜测路径。 - 第一次为项目 ingest 持续记忆时，如果用户没有告知 spec 路径或 RTL 路径，必须询问用户；不要自己猜测 spec/RTL 根目录。仅创建空 wiki 骨架不要求提供这些路径。
- 需要校验格式时运行：
python <xverif-root>/skills/xwiki/scripts/validate_xwiki.py- 查询 wiki 可隐式触发；创建、修改、废弃或写回页面必须有用户明确要求、项目规则授权或任务范围授权。未授权时只在最终结果中列出建议写回的稳定结论。
- 初始化空 wiki 目录时优先运行：
python <xverif-root>/skills/xwiki/scripts/init_xwiki.py --wiki-dir "$XWIKI_DIR" --validate何时使用
- 需要了解验证环境、DUT 功能、接口、reset/clock、memory map、interrupt、testbench、agent、sequence、checker、scoreboard、coverage、workflow、debug 入口。
- 需要把源码、README、spec、test、wave/debug 报告或用户说明编译进持久项目记忆。
- 需要查询、创建、修改或废弃验证项目 wiki 页面。
- 需要检查 wiki 格式是否能被后续 agent 可靠读取。
- 获得写回授权且 debug 完一个 case fail 后，把稳定结论记录到
de_issue或dv_issue描述对象中。
查询顺序
- 从
$XWIKI_DIR/index.md开始，找候选主题和入口页面。 - 按描述对象进入
$XWIKI_DIR/de/、dv/、de_issue/或dv_issue/，先读该目录及沿途子目录的index.md和log.md。 - 读取相关 concept 页面，看 frontmatter、正文、出链、evidence 和 citations。
- 如果存在
_index/backlinks.md或_index/tags.md，用它们做反向或 tag 索引。 - 如果仍找不到，只在
$XWIKI_DIR内用rg搜索 Markdown。 - 搜索命中后必须回到页面级证据；不要只引用 grep 片段作答。
- wiki 信息不足时再读 raw source；只有已获得写回授权时才把稳定发现编译回 wiki，未授权时在最终结果列出建议写回的页面和结论。
详细查询和 CRUD 规则见 references/wiki-crud.md。
编译要求
严格执行 LLM Wiki 编译过程：raw sources 是事实来源，wiki 是编译产物，schema 由本 skill 规定。每次 ingest/update 必须读旧页面、抽取新事实、合并或新增 concept、更新 index/反向索引、追加 log、运行校验并汇报 unknowns。
验证层次 prompt 生命周期
prompt 的加载范围由当前操作阶段决定，不得在每次 xwiki 查询或写回时无条件读取同层全部 prompt：
- 空骨架初始化：只运行
init_xwiki.py创建目录和 scaffold，不读取 topic prompt，也不要求确定验证层次。 - 首次项目 ingest：先确定当前验证层次是
bt、it、st还是soc。验证层次可来自用户说明、wiki index、已有 concept、仓库目录或验证环境命名；如果无法确定，必须询问用户，不得猜测。确定后读取该层次references/prompts/<level>/prompts/*.md下的全部 prompt，并创建$XWIKI_DIR/_index/prompt-profile.md，持久记录验证层次、初始化时间、已读取的 prompt 文件集合、topic 到 prompt 文件的映射和是否需要刷新。profile 不复制 prompt 正文。 - 普通查询：不读取任何 topic prompt；只按查询顺序读取 wiki，必要时再读 raw source。
- 增量 concept 更新：先读取
prompt-profile.md，只读取目标 topic 对应的一个或多个 prompt，并结合 references/prompt-output-requirements.md 生成或更新页面。不得为了更新单个 topic 重新读取同层全部 prompt。 - issue、index 或 log 更新：不读取 topic prompt；按 wiki schema、CRUD 和证据规则直接维护。
- 全量刷新：只有 prompt profile 缺失、验证层次发生变化或用户明确要求全量重建/刷新时，才重新读取当前层次的全部 prompt 并重建 profile。
首次 ingest、增量 concept 更新或全量刷新完成后，汇报验证层次、实际读取的 prompt 集合和主要使用的 prompt；没有读取 prompt 时不需要汇报 prompt 集合。
Show full SKILL.md (213 more words)Show less
Wiki 描述对象分类硬规则
所有 wiki Markdown 文件的 frontmatter 必须包含 object_type，且只能使用 de、dv、de_issue、dv_issue 四个值之一。该字段用于区分页面描述对象，不等同于页面标题、tag 或验证层次。
de：描述设计实现、RTL、接口、微架构、协议行为、设计参数和数据路径。dv：描述验证环境、sequence、checker、scoreboard、coverage、test、debug workflow 和仿真入口。de_issue：持续记录设计、RTL、spec、协议定义、性能需求或微架构侧问题和风险。dv_issue：持续记录验证环境、RM、checker、scoreboard、sequence、testbench、脚本、配置或 DV 假设侧问题和风险。
根 index.md 默认使用 object_type: dv；每个目录的 index.md 使用该目录对应的 object_type，de_issue/spec/index.md 和 de_issue/rtl/index.md 使用 object_type: de_issue。log.md 默认使用所在目录对应的 issue 类型：de/ 下使用 de_issue，dv/ 下使用 dv_issue，de_issue/ 及其子目录使用 de_issue，dv_issue/ 下使用 dv_issue。获得写回授权后，debug 完 case fail 时更新 de_issue 或 dv_issue 页面；根因未完全确认时，也要作为候选 issue 记录 unknown 和下一步证据。
禁止把具体仿真产物作为 wiki evidence 或 citation，例如单次 run 的 FSDB/VCD、simv 产物、临时日志、coverage 临时目录、scratch 报告、<repo>/tmp 文件。这些不会进入 git，只能作为当次 debug 的 raw observation；写入 wiki 时必须编译成稳定结论，并引用可追踪的 spec、RTL、test、脚本、README 或已提交文档。
获得写回授权后，case fail debug 结束时写回 wiki：
- 设计、RTL、spec、协议定义、性能需求或微架构侧问题写入
de_issue。 - testbench、UVM env、RM、checker、scoreboard、sequence、配置、脚本、仿真参数或 DV 假设问题写入
dv_issue。 - 如果根因未完全确认，仍要更新对应候选 issue 页面，把结论标为未确认，并记录下一步需要的证据。
写入或更新任何具体 de_issue 或 dv_issue 页面前，必须检查本次分析涉及的每个 RTL source root 是否属于 Git 仓库。具体 issue 页面的 frontmatter 必须包含 rtl_revisions：已确认没有相关 RTL Git 仓库时写 []；存在时逐仓库记录稳定逻辑名称、HEAD 的 40 位 commit、所有精确指向该 commit 的 tag，以及包含未跟踪文件在内的 dirty 状态。相关 submodule 作为独立仓库记录。tag 不能替代 commit；仓库名称不得使用本机绝对路径。无法取得版本信息时不得写空列表或猜测值，必须向用户报告阻塞并询问。
根因主题与存储对象固定映射：env_bug 写入 dv_issue/，rtl_bug 写入 de_issue/rtl/，spec_bug 写入 de_issue/spec/。主题名用于问题分类，frontmatter 的 object_type 仍只能使用 dv_issue 或 de_issue，不能写成 env_bug、rtl_bug、spec_bug。
详细流程见 references/compile-process.md。
Wiki 格式
wiki 根目录必须包含 index.md、de/、dv/、de_issue/、dv_issue/。根目录不再要求 log.md；各描述对象目录必须包含自己的 index.md 和 log.md。de_issue/ 下必须再区分 spec/ 与 rtl/，且二者也必须各自包含 index.md 和 log.md。
wiki 允许多层子目录，但 $XWIKI_DIR 下除 _index/、archive/、deprecated/ 以外，任何包含 Markdown 页面或子目录的目录都必须同时包含 index.md 和 log.md，形成可局部查询、可局部追溯的分层记忆。所有 Markdown 文件都必须有 YAML frontmatter，至少包含 type、title、description、object_type。object_type 只能是 de、dv、de_issue、dv_issue。具体 issue 页面还必须包含合法的 rtl_revisions。链接必须相对可解析，禁止本机绝对路径和 file://。
创建空 wiki 骨架时使用 scripts/init_xwiki.py，不要手写目录骨架。脚本默认保留已有文件；需要重建脚本管理的 index/log/_index 文件时显式传 --force；只查看计划时传 --dry-run。
完整格式见 references/wiki-spec.md。
验证 topic prompts
BT/IT/ST/SoC prompt 保存在 references/prompts。首次项目 ingest 和全量刷新读取对应层次 prompts/ 目录下的全部 Markdown 文件；后续增量 concept 更新只读取 prompt-profile.md 映射到目标 topic 的 prompt。普通查询及 issue、index、log 更新不读取 topic prompt。
© BLANK2077, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 68 other files (scripts, references) in skills/xwiki of BLANK2077/xverif.
- SKILL.md
- agents/openai.yaml
- references/compile-process.md
- references/prompt-output-requirements.md
- references/prompts/bt/prompts/backpressure.md
- references/prompts/bt/prompts/checker.md
- references/prompts/bt/prompts/credit.md
- references/prompts/bt/prompts/debug.md
- references/prompts/bt/prompts/design_overview.md
- references/prompts/bt/prompts/fifos.md
- references/prompts/bt/prompts/interfaces.md
- references/prompts/bt/prompts/ordering.md
- references/prompts/bt/prompts/osd.md
- references/prompts/bt/prompts/project.md
- references/prompts/bt/prompts/risks.md
- references/prompts/bt/prompts/rm.md
- … and 53 more
Open the folder on GitHubat commit 6aa13f9
Compare with similar skills
Xwiki next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Xwiki this skillBLANK2077/xverif | 108 | — | ~1.4k | Automated safety check: Pass | MIT | |
| Karpathy LLM WikiAstro-Han/karpathy-llm-wiki | 2.4k | — | ~3.6k | Automated safety check: Pass | MIT | |
| LLM Wikilewislulu/llm-wiki-skill | 655 | — | ~3.7k | Automated safety check: Pass | None | |
| Wiki Builderrohitg00/pro-workflow | 2.9k | — | ~1k | Automated safety check: Pass | None | |
| Codex History IngestAr9av/obsidian-wiki | 3.5k | — | ~2.2k | Automated safety check: Notes | MIT | |
| Arkon Editnduckmink/arkon | 1.5k | — | ~1.6k | Automated safety check: Pass | Custom licence |
Similar skills
LLM Wiki
lewislulu/llm-wiki-skill
Build and maintain a Karpathy-style LLM knowledge base — a self-compiling Obsidian markdown wiki where an Agent ingests raw sources, compiles cross-linked concept/entity/summary pages, answers…
Wiki Builder
rohitg00/pro-workflow
Start, structure, and grow a persistent research wiki indexed in pro-workflow's SQLite knowledge base.
Codex History Ingest
Ar9av/obsidian-wiki
Ingest Codex CLI conversation/session history into Obsidian as distilled knowledge.
Arkon Edit
nduckmink/arkon
Propose or directly apply edits to Arkon wiki pages, including proposing brand new pages.
Research Wiki Builder
dair-ai/dair-academy-plugins
Creates and maintains configurable research wikis: scaffold a folder, add sources, compile pages and indexes, and file query answers back.
More from BLANK2077/xverif
X Npi
BLANK2077/xverif
当 AI agent 需要使用 Synopsys pynpi 编写 Python 脚本，进行批量 FSDB 波形统计、值扫描、APB/AXI/valid-ready stream 协议分析、VCS/Verdi coverage database 查询，或静态设计 driver/load 查询时使用。离线大规模分析脚本和报告优先使用本 skill；xdebug 风格的实时…
Xsimdebug
BLANK2077/xverif
当 AI agent 需要定位 SystemVerilog/UVM 验证问题、重新编译代价过高或不允许再次编译，并需要直接通过终端 PTY 使用 Synopsys VCS UCLI 或 Cadence Xcelium/Xrun Tcl 实时调试正在运行的仿真时使用，包括断点、Ctrl-C 暂停、运行时变量查询和控制流。若已有仿真 log 已足以定位问题，则不使用本 skill。
Xverif
BLANK2077/xverif
用于芯片验证中的确定性事实查询和计算：daidir/FSDB debug、coverage、 SystemVerilog bit 计算、entry 解码、日志位置恢复、SVA 解释和波形渲染。
Xverif Admin
BLANK2077/xverif
用于 xverif 的安装配置、MCP direct/LSF backend、ssh 远端 MCP（mcpssh）、 SDK-free LSF CLI、UDS/TCP/file transport、session tombstone/gc、timeout、 环境变量、license 和 server 启动排障。普通波形、coverage、bit 或协议查询使用 xverif。
Related
Categories
Questions about Xwiki
What does Xwiki do?
当 AI agent 需要在芯片验证任务中复用持续记忆时使用：查询或维护由 XWIKIDIR 指向的验证项目 LLM wiki，了解验证环境、DUT 功能、 接口、testbench、sequence、checker、coverage、workflow、debug 入口、 BT/IT/ST/SoC 项目上下文，并在用户明确要求、项目 AGENTS.md 授权或任务明确…. Xwiki is an agent skill from BLANK2077/xverif.
When should I use Xwiki?
Xwiki fits situations like: tasks that involve LLM wikis.
How do I install Xwiki in Claude Code?
Run `npx skills add BLANK2077/xverif --skill xwiki -a claude-code`. Or copy the skill folder (skills/xwiki in BLANK2077/xverif) into .claude/skills/xwiki in your project. Claude Code loads it when a task matches its description.
How do I install Xwiki in Codex?
Run `npx skills add BLANK2077/xverif --skill xwiki -a codex`. Or copy the skill folder (skills/xwiki in BLANK2077/xverif) into .agents/skills/xwiki in your project. Codex loads it when a task matches its description.
Can I use Xwiki in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add BLANK2077/xverif --skill xwiki -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/xwiki, .gemini/skills/xwiki, .github/skills/xwiki and .opencode/skills/xwiki in your project.
What does Xwiki need to run?
Going by SKILL.md and its folder, Xwiki needs the command-line tools its instructions call (python). Our summary lists: Python 3.
Does Xwiki access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Xwiki safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
What licence does Xwiki use?
Xwiki is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Xwiki use?
About 1.4k tokens (SKILL.md is roughly 5.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 12k tokens, read only when the agent opens those files.
What are the alternatives to Xwiki?
Skills that share tags, products or a category with Xwiki: Karpathy LLM Wiki (Astro-Han/karpathy-llm-wiki, 2.4k stars), LLM Wiki (lewislulu/llm-wiki-skill, 655 stars), Wiki Builder (rohitg00/pro-workflow, 2.9k stars) and Codex History Ingest (Ar9av/obsidian-wiki, 3.5k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Xwiki?
BLANK2077 (a GitHub user) maintains it in BLANK2077/xverif, which has 108 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on October 8, 2026.
Source: BLANK2077/xverif on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.