GITHUB AGENT SKILLS / 技能收录

给你的 AI 助手
装上专业技能

我们从 GitHub 持续收录、审核并分类优质的 Agent Skills,按职业与使用场景整理,复制一行命令即可安装到 Claude Code、Cursor 等智能体。

01 /

按职业浏览 Agent Skills

为不同岗位挑选的高频技能

全部职业
02 /

按使用场景浏览 Agent Skills

从你要完成的任务出发

全部场景
03 /

按创作者浏览 Agent Skills

按 GitHub Stars 排序的优秀创作者

全部创作者

全部 Skills 482556

AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-sync-specs

Ensure a component implementation is synchronized with its specifications.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-sweep

Fast breadth-first TRIAGE across many Certus components — property inventory and spec-vs-code divergence list only, no proving. Produces the three property YAMLs per component (the same files `component-verify` Step 2 consumes) plus a cross-component divergence report, in hours rather than the days a full prover pass costs. Use it to decide WHERE to spend prover time, and to get the property denominators for every component. It never proves, never scores, never touches a branch. It is NOT a substitute for `component-verify` and produces no scoring page.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-repair

Repair the code behind a verification finding — a level-2 bug, or on request a level-1 spec<->code discordance — completely and across components if needed, and open a PR a non-expert can approve from the evidence. Runs the vendored repair-agent; acceptance is decided by repair_accept_repo.sh, never by the agent.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-remove

Remove a component from the code base.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-poison

Deliberately introduce a controlled number of specification/code misalignments ("poisons") into a single Certus component on a throwaway branch, in order to test translation-verification tooling such as `component-check-spec-translation`. Each poison intentionally desynchronizes the component's `spec.md` from its implementation while keeping the code compiling. Use only for red-team / evaluation of the audit skills, never on a branch intended to merge.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-poison-verif

Deliberately introduce a controlled number of concrete-code/abstract-code misalignments ("verif poisons") into a single Certus component on a throwaway branch, in order to test formal-verification drift tooling (Creusot mirror/equality checks, Kani assume-audits, and `tools-spin-sync`). Each poison intentionally desynchronizes the component's verification artifacts (a `verif/` Creusot mirror, a `#[cfg(kani)] mod verification`, or a Spin `.pml` model) from the shipped implementation while keeping the code compiling. Use only for red-team / evaluation of the verification skills, never on a branch intended to merge.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-poison-dual

暂无简介

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-new-version

Create a new version of a component

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-make-new

Create a new component skeleton

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-make-new-actor

Create a new actor component skeleton

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-like-existing

Bootstrap a new component from an existing one, preserving its provided interfaces and receptacles under a different component name.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-check-verif-translation

Cross-check a Certus component's abstract verification artifacts (a `verif/` Creusot mirror, a `#[cfg(kani)] mod verification`, or a Spin `.pml` model) against its concrete shipped implementation, then summarize concrete-code/abstract-code drift with source evidence. Use when auditing whether a formal-verification harness still faithfully mirrors the code it claims to prove, when checking for vacuous contracts or unmatched assumptions after code changes, or when validating what `component-poison-verif` seeded. This is a read-only audit and does not reconcile, repair, or edit either side; it does not run spec/model sync or repair skills.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-check-spec-translation

Cross-check a Certus component's nested Spec Kit `specs/**/spec.md` specifications against its generated implementation and tests, then summarize semantic translation mismatches with source evidence. Use when auditing whether generated code faithfully implements component requirements, acceptance scenarios, interfaces, defaults, constraints, errors, or success criteria; when checking implementation drift after generation or code changes; or when asked to compare a component spec with code. This is a read-only audit and does not reconcile or edit either side.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-check-leakage

Checks that any component that is being used by another component is only accessed through the component's interface and does not interact directly with the struct.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

build-property-inventory

Build the ONE tool-independent, source-traced verifiable-property inventory for a component and bundle every property onto the I<component> public methods that implement it. Reconciles two independent, BLIND per-artifact extractions (spec and code, isolated from each other) into a single ID-keyed set with owning-method bundles and counts. Does NOT name a prover and does NOT record proof status — that is added downstream. This is Role 1 (the single source of truth) that `tools-verify-*-with-properties` and `tools-aggregate-coverage-by-interface` consume.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-poison

Deliberately introduce a controlled number of specification/code misalignments ("poisons") into a single Certus component on a throwaway branch, in order to test translation-verification tooling such as `component-check-spec-translation`. Each poison intentionally desynchronizes the component's `spec.md` from its implementation while keeping the code compiling. Use only for red-team / evaluation of the audit skills, never on a branch intended to merge.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-check-verif-translation

Cross-check a Certus component's abstract verification artifacts (a `verif/` Creusot mirror, a `#[cfg(kani)] mod verification`, or a Spin `.pml` model) against its concrete shipped implementation, then summarize concrete-code/abstract-code drift with source evidence. Use when auditing whether a formal-verification harness still faithfully mirrors the code it claims to prove, when checking for vacuous contracts or unmatched assumptions after code changes, or when validating what `component-poison-verif` seeded. This is a read-only audit and does not reconcile, repair, or edit either side; it does not run spec/model sync or repair skills.

★ 3
AI-native-Systems-Research
AI-native-Systems-Research/ai-native-storage-certus

component-check-spec-translation

Cross-check a Certus component's nested Spec Kit `specs/**/spec.md` specifications against its generated implementation and tests, then summarize semantic translation mismatches with source evidence. Use when auditing whether generated code faithfully implements component requirements, acceptance scenarios, interfaces, defaults, constraints, errors, or success criteria; when checking implementation drift after generation or code changes; or when asked to compare a component spec with code. This is a read-only audit and does not reconcile or edit either side.

★ 3