haoxiang-xu/PuPu

PuPu is a lightweight, cross-platform desktop AI client that works with both local and cloud-hosted models. Whether you prefer running models on your own machine or connecting to providers like OpenAI and Anthropic, PuPu gives you a unified, elegant interface — your AI, your rules.

This repository also configures its own agents. See what PuPu tells them →

36Stars on the repository
21Mods indexed here, across every type
todayLast push, which is what freshness is scored on
Apache-2.0Licence, which decides whether bodies are shown

gitnexus-cli

01

haoxiang-xu/PuPu

Skill Claude Code needs its repo

Use when the user needs to run GitNexus CLI commands like analyze/index a repo, check status, clean the index, generate a wiki, or list indexed repos. Examples: "Index this repo", "Reanalyze the codebase", "Generate a wiki".

not rated 36 today A 59 tokens copy · 89% Apache-2.0

gitnexus-debugging

02

haoxiang-xu/PuPu

Skill Claude CodeCodex

Use when the user is debugging a bug, tracing an error, or asking why something fails. Examples: "Why is X failing?", "Where does this error come from?", "Trace this bug".

not rated 36 today A SkillSpector: pass 46 tokens copy · 100% Apache-2.0

gitnexus-exploring

03

haoxiang-xu/PuPu

Skill Claude CodeCodex

Use when the user asks how code works, wants to understand architecture, trace execution flows, or explore unfamiliar parts of the codebase. Examples: "How does X work?", "What calls this function?", "Show me the auth flow".

not rated 36 today A SkillSpector: pass 55 tokens copy · 100% Apache-2.0

gitnexus-guide

04

haoxiang-xu/PuPu

Skill Claude CodeCodex

Use when the user asks about GitNexus itself — available tools, how to query the knowledge graph, MCP resources, graph schema, or workflow reference. Examples: "What GitNexus tools are available?", "How do I use GitNexus?".

not rated 36 today A 56 tokens copy · 100% Apache-2.0

haoxiang-xu/PuPu

Skill Claude CodeCodex

Use when the user wants to know what will break if they change something, or needs safety analysis before editing code. Examples: "Is it safe to change X?", "What depends on this?", "What will break?".

not rated 36 today A SkillSpector: warn 50 tokens copy · 91% Apache-2.0

haoxiang-xu/PuPu

Skill Claude CodeCodex

Use when the user wants to rename, extract, split, move, or restructure code safely. Examples: "Rename this function", "Extract this into a module", "Refactor this class", "Move this to a separate file".

not rated 36 today A SkillSpector: pass 53 tokens copy · 100% Apache-2.0

haoxiang-xu/PuPu

Skill Codex

A release-closing procedure for a PuPu version in GitHub Projects. It checks one release issue and its direct sub-issues, then records whether the release evidence is complete.

not rated 36 changed 1mo ago A SkillSpector: pass 70 tokens original Apache-2.0

release-open-sprint

08

haoxiang-xu/PuPu

Skill Codex

A workflow for opening a new PuPu software release as a parent GitHub issue, with its release work tracked in smaller child issues. It treats the GitHub Project as the official record of the release scope.

not rated 36 changed 1mo ago A SkillSpector: pass 87 tokens original Apache-2.0

haoxiang-xu/PuPu

Skill Codex

Analyze PuPu's growth, GitHub traffic, installer downloads, release performance, community and contributor health; produce a founder-facing weekly COO report or a light release-planning baseline. Use for growth analysis and reporting; route GitHub tags changes to community-repo-tags-optimization.

not rated 36 today A 62 tokens original Apache-2.0

haoxiang-xu/PuPu

Skill Claude CodeCodex

A workflow for measuring and changing the topics attached to a GitHub repository. GitHub topics are labels that help people find a repository, and this workflow records traffic and download data before and after changes.

not rated 36 changed 1mo ago A 97 tokens original Apache-2.0

haoxiang-xu/PuPu

Skill Claude CodeCodex

Produce version-bound PuPu security verification evidence for a selected Skill, MCP server or plugin, including code, prompt, dependency and isolated behavior checks. Use for verifying or re-verifying marketplace candidates, reviewing a verification report, or assessing badge revocation; ordinary catalog edits do not…

not rated 36 today A 65 tokens original Apache-2.0

haoxiang-xu/PuPu

Skill Codex

Standardize modal UI to match the Settings/Tools baseline used in this repo. Use when a user asks to align modal style/spacing/typography/close-button behavior, or says a modal looks inconsistent (for example close button position, border radius, title size/font, or empty-state panel visibility).

not rated 36 today A 66 tokens original Apache-2.0

issue-draft-ticket

13

haoxiang-xu/PuPu

Skill Codex

A workflow for turning a short request into a draft PuPu GitHub issue under an open release. The draft records the intended outcome and outline-level acceptance points, rather than pretending to be a complete specification.

not rated 36 today A 70 tokens original Apache-2.0

issue-feature-audit

14

haoxiang-xu/PuPu

Skill Codex

A consistency audit for a completed PuPu feature before its ticket is marked done. It checks the feature and its surrounding impact, and can also scan the whole project for missing translation keys.

not rated 36 today A 94 tokens original Apache-2.0

issue-refine-ticket

15

haoxiang-xu/PuPu

Skill Codex

Use when a PuPu release sub-issue needs a researched, zero-context implementation brief — "refine #123" or when an implementation agent wants to expand a [DRAFT]. Optional support, not a gate: the agent implementing the ticket may refine its body directly.

not rated 36 today A 60 tokens original Apache-2.0

haoxiang-xu/PuPu

Skill Claude CodeCodex needs its repo

Operate PuPu's protected GitHub Actions release chain when the project owner asks to build, qualify, stage, publish, resume, or inspect a release candidate. Use for exact run orchestration; do not use for release scope decisions or feature audits.

not rated 36 today A 55 tokens original Apache-2.0

haoxiang-xu/PuPu

Skill Claude CodeCodex

An evidence-based review process for long-running language-model test reports and task artifacts, where a harness is the system that runs benchmark tasks.

not rated 36 today A 62 tokens original Apache-2.0

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: