Official agent skill

Remediate Image Cves

by kubernetes-sigs in kubernetes-sigs/cloud-provider-azure

Orchestrate end-to-end CVE remediation for the Linux CCM, CNM, and health-probe-proxy images on cloud-provider-azure master or a release-X.Y branch, including builds, repeated Trivy verification…

OfficialApache-2.0Auto-check passedSecurity

Install Remediate Image Cves

skills CLI
$ npx skills add kubernetes-sigs/cloud-provider-azure --skill remediate-image-cves -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install kubernetes-sigs/cloud-provider-azure remediate-image-cves --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/kubernetes-sigs/cloud-provider-azure.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/remediate-image-cves .claude/skills/remediate-image-cves && rm -rf skills-src

Use ~/.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/

Facts

Skill name
remediate-image-cves
GitHub stars
294
Token cost
~3.9k tokens
SKILL.md length
2,126 words
Files
2 (incl. assets)
Skills in repo
12
Repo updated
First seen
Licence
Apache-2.0

At a glance

Orchestrate end-to-end CVE remediation for the Linux CCM, CNM, and health-probe-proxy images on cloud-provider-azure master or a release-X.Y branch, including builds, repeated Trivy verification…

  • Works in 6 steps: Require a clean worktree before changing… → Fail fast if Git, GitHub CLI, Trivy,… → Require both git var GIT_AUTHOR_IDENT and → …
  • The user wants to verify
  • SKILL.md covers Sources of Truth, Git Lock Discipline, Preconditions and Branch and Image Identity, plus 4 more sections
  • Calls git, go and make

What it does

Remediate Image Cves is an agent skill from kubernetes-sigs/cloud-provider-azure, published by the product's own GitHub organization. Orchestrate end-to-end CVE remediation for the Linux CCM, CNM, and health-probe-proxy images on cloud-provider-azure master or a release-X.Y branch, including builds, repeated Trivy verification, checkpoint commits, cleanup, validation, push, and pull-request creation. Use when the user wants to verify or fix CVEs on master or a release branch, run the three-image CVE workflow, or open a CVE remediation PR for either supported target branch.

Its SKILL.md is about 3.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files, including assets (for example `assets/pull-request-template.md`).

It sits in Security, covering Vulnerability scanning. It works with Kubernetes, Microsoft Azure, Trivy and Linux. The repository describes itself as: Cloud provider for Azure. The licence is Apache-2.0.

When your agent uses it

  • The user wants to verify
  • Fix CVEs on master
  • A release branch
  • Run the three-image CVE workflow

Example prompts

  • “/remediate-image-cves”

Requirements

  • Docker

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

  1. Require a clean worktree before changing branches. Preserve all pre-existing
  2. Fail fast if Git, GitHub CLI, Trivy, Docker with Buildx or Podman with native
  3. Require both git var GIT_AUTHOR_IDENT and
  4. Require upstream to contain the input branch and verify, before any build,
  5. Check for an existing open CVE-remediation pull request targeting the input
  6. If this worktree already has unfinished fix-image-cves state, stop and

What it can do on your machine

Read from SKILL.md and the folder at commit 0201852. 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

    Shell commands in SKILL.md call:

    • git
    • go
    • make
    • kind

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    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

Remediate Image Cves loads about 3.9k tokens when it runs. Until then it costs about 117 tokens; SKILL.md has 2,126 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~117
When it runs · the whole SKILL.md, loaded when a task matches
~3.9k

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); files beside SKILL.md are not scanned.

SKILL.md

The full file from kubernetes-sigs/cloud-provider-azure at commit 0201852, republished under its Apache-2.0 licence (© kubernetes-sigs). 2,126 words, ~3,914 tokens.

Download SKILL.mdSave it as .claude/skills/remediate-image-cves/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
remediate-image-cves
description
Orchestrate end-to-end CVE remediation for the Linux CCM, CNM, and health-probe-proxy images on cloud-provider-azure master or a release-X.Y branch, including builds, repeated Trivy verification, checkpoint commits, cleanup, validation, push, and pull-request creation. Use when the user wants to verify or fix CVEs on master or a release branch, run the three-image CVE workflow, or open a CVE remediation PR for either supported target branch.

Remediate Image CVEs

Remediate actionable fixable vulnerabilities in the Linux CCM, CNM, and health-probe-proxy images for one input target branch.

Require the caller to provide an input branch that is exactly master or matches ^release-[0-9]+\.[0-9]+$. Operate only on the current checkout; callers may run this skill concurrently in separate worktrees for separate input branches.

Sources of Truth

Read these repository skills before taking any action:

  • .agents/skills/build-images/SKILL.md
  • .agents/skills/fix-image-cves/SKILL.md
  • .agents/skills/sync-go-modules/SKILL.md

Follow those skills as the sources of truth for Makefile image builds and Trivy-based CVE remediation and for final repository-wide Go module consistency. Do not duplicate, override, or extend their commands. This skill supplies orchestration, lifecycle, Git, validation, cleanup, and pull-request policy only. If a dependency is missing or conflicts with this skill, stop and report the blocker.

Before checking out the input branch, copy this complete skill directory and all three dependency skill directories into a unique run-scoped temporary directory outside the repository. Use that exact tooling snapshot for the entire run; do not depend on skill files present on the target branch. Invoke both image helpers and the module-sync helper with their explicit --repo input pointing at the current worktree. Remove the temporary tooling snapshot during final cleanup.

Git Lock Discipline

Run every orchestration-owned read-only Git inspection with command-scoped GIT_OPTIONAL_LOCKS=0. This includes status, diff, log, show, rev-parse, ls-files, ls-tree, and ref checks. The image-CVE and module-sync helpers enforce the same rule internally for their read-only Git subprocesses.

Do not export this setting across the workflow or pass it to mutating Git commands. Fetch, checkout, branch or ref updates, staging, commits, and pushes must run serially with normal required locking. Before each mutation and after each long-running helper exits, inspect the current worktree Git directory for lock files. If an unexpected lock exists, stop and report its path, owner, size, modification time, and holder check when available. Never delete, rename, bypass, or work around it as part of this skill.

Preconditions

  1. Require a clean worktree before changing branches. Preserve all pre-existing work; never reset, discard, or overwrite it.
  2. Fail fast if Git, GitHub CLI, Trivy, Docker with Buildx or Podman with native build support, or a local Go version compatible with the checked-out branch is unavailable. Before each CVE apply, also require an available local Go version compatible with the fresh plan's Go directive actions. Do not install or upgrade host software or enable automatic toolchain downloads.
  3. Require both git var GIT_AUTHOR_IDENT and git var GIT_COMMITTER_IDENT to succeed before any build so checkpoint commits cannot fail after remediation has started.
  4. Require upstream to contain the input branch and verify, before any build, that the current authenticated account can push to origin and open a pull request against upstream. Check repository permissions rather than hard-coding classic-token scope names.
  5. Check for an existing open CVE-remediation pull request targeting the input branch. If one exists, stop and report its URL instead of creating a duplicate.
  6. If this worktree already has unfinished fix-image-cves state, stop and report it instead of deleting evidence from an earlier run.

Branch and Image Identity

Validate that the input branch is exactly master or matches ^release-[0-9]+\.[0-9]+$. Fetch and check out its current tip from upstream, using only a fast-forward update if a local branch already exists. Never hard-reset a local branch; stop if it cannot be fast-forwarded safely. Then create a branch named cve-fix-<input-branch>-<UTC timestamp>, for example cve-fix-master-20260721T013000Z or cve-fix-release-1.36-20260721T013000Z.

Use the build skill with:

  • image registry: local
  • image tag: cve-<short-sha>-<UTC timestamp>-<run-id>
  • explicit make inputs: ARCH=amd64 and OUTPUT_TYPE=docker
  • inherited make inputs removed: OUTPUT_FLAG and BUILDX_EXTRA_FLAGS

Generate <run-id> once as a lowercase UUID without hyphens and retain it for the complete run. The tag must be unique to the run so concurrent worktrees cannot overwrite, scan, or delete one another's verification images. Pass those make inputs through the build skill's --set and --unset options on every build, including rebuilds and final verification builds. These images are disposable and must never be pushed.

Select one container runtime for the complete run. Respect an explicit CONTAINER_CLI; otherwise use the Makefile preference of Podman when available and Docker otherwise. Resolve and record its absolute path, then pass it through the build skill as --set CONTAINER_CLI=<path> on every build. Do not build with one runtime and scan or clean up with another.

After each build, use the selected runtime to resolve and record the canonical local image reference. Use that exact reference for Trivy scans, rescans, and cleanup. Podman may normalize local/... to localhost/local/...; do not assume the Makefile reference is also the canonical runtime reference.

Verification Set

Use this order and identity in every phase:

ImageBuild aliasMakefile image referenceModule rootDockerfile
CCMccmlocal/azure-cloud-controller-manager:<tag>.Dockerfile
CNMcnmlocal/azure-cloud-node-manager:<tag>-linux-amd64.cloud-node-manager.Dockerfile
health-probe-proxyhpplocal/health-probe-proxy:<tag>health-probe-proxyhealth-probe-proxy/Dockerfile

For every build, allow concurrent builds from other worktrees. Docker builds share a host-level Buildx builder; Podman uses its native builder. Pass the build helper's --retry-transient-runtime-errors option on every build. The build skill owns runtime-specific failure classification, health checks, bounded delay, exact retry execution, and the two-attempt limit. Do not run an additional outer retry. If the helper still returns a failure, preserve both attempts in the run summary and stop.

Baseline Phase

Establish the three-image baseline before mutating source:

  1. Build all three images from the untouched input-branch source before running any fix-image-cves apply command.
  2. For each baseline image, run scan and plan with its canonical image reference, module root, and Dockerfile. Record the complete plan, actionable findings, unsupported fixable findings, and residual risks outside the helper state before scanning the next image.
  3. Treat Go toolchain findings and findings without a FixedVersion as residual risks. Treat any OTHER finding with a non-empty FixedVersion as an unsupported fixable finding and stop without applying changes or opening a PR, preserving that image's state.
  4. The helper has one worktree-local state file. When continuing, run its clean command after recording each successful baseline result so the next image starts with empty state. If a baseline scan or plan fails, preserve that image's state and stop.

If all three baseline plans contain zero Go-module actions, zero base-image actions, and zero unsupported fixable findings, exit successfully without committing, pushing, or opening a PR. Only residual findings may remain on this no-change path.

Remediation Phase

After the complete baseline is recorded, process images sequentially in the table order:

  1. Rebuild the image from the current source and resolve its canonical runtime reference. Run a fresh scan and plan; do not apply the saved baseline plan because an earlier image may already have changed a shared module.
  2. Review every fresh plan. If it contains Go directive actions, select a locally installed Go version at least as new as the highest target before apply, and keep GOTOOLCHAIN=local. Apply Go-module and directive fixes through fix-image-cves, including compatible pinned Go builder targets for every affected verification Dockerfile (both CCM and CNM for the root module). Follow the fix skill's Dockerfile:builder target policy; keep already-compatible builders unchanged. For a fixable runtime base-image finding, automatically replace only the digest of the existing registry, repository, and tag. If runtime base-image remediation would change the image family or tag, stop for review. Stop and report any other permission, judgment, or conflict-resolution requirement instead of guessing.
  3. If the fresh plan contains no Go-module or base-image actions and no unsupported fixable findings, record its residual risks, run clean, and continue to the next image. Skip apply, file verification, remediation rebuild, and planned-key rescan for this zero-action plan.
  4. If the fresh plan has actions, run apply, then file verification. Rebuild the image and run verify --rescan to prove the planned vulnerability keys disappeared.
  5. Before starting a new scan, record the apply state's exact modified-file list and verification results; scan replaces prior apply and verify state. Then run a fresh scan and plan of the rebuilt image to detect newly introduced actionable or unsupported fixable findings.
  6. After the fresh plan is reviewed, commit only the recorded modified files as the checkpoint for that cycle. Do not commit immediately after file verification. The fresh plan need not be empty for an intermediate cycle. Never reset or revert an earlier checkpoint, and keep all checkpoint commits in the final PR.
  7. If the fresh plan has no remaining actions or unsupported fixable findings, record residual risks, run clean, and continue to the next image. If it has more actions, use that current plan for the next cycle. Allow at most three apply/verify/fresh-plan cycles per image.
  8. If verification fails, an unsupported fixable finding appears, or actions remain after the third cycle, mark the run incomplete. Stop without pushing or opening a PR and preserve the current helper state, source changes, and local checkpoint commits for diagnosis.
Show full SKILL.md (673 more words)Show less

Final Gate

After all remediation cycles:

  1. From the clean checkpointed source, run sync-go-modules once to normalize every tracked module and the root vendor tree exactly as CI does. If the vendor dependency set changes, rerun it with --allow-dirty and --update-vendor-licenses, then run it once more with --allow-dirty to normalize the root module after license generation. Inspect the resulting diff and commit only expected go.mod, go.sum, vendor/, and LICENSES/ paths as a final module-consistency checkpoint. Never absorb unrelated files.
  2. Run sync-go-modules --check-clean. Do not continue unless it produces no diff. If vendor-license generation was needed, normalize modules again after that generation before running --check-clean because the upstream license tooling executes go list -m all.
  3. Run the full repository unit-test suite with make test-unit.
  4. Run git diff --check against the input branch.
  5. Rebuild all three images from the final committed source.
  6. Freshly scan and plan all three rebuilt images in table order. Record each result before scanning the next image; a new scan replaces the previous state, so no intermediate clean is required. Stop immediately on failure so the failing image's current state remains available. Verification is complete only when none has remaining Go-module or base-image actions and none has an unsupported fixable OTHER finding. Preserve findings without a fixed version and Go toolchain findings as residual risks.
  7. Require a clean worktree containing only the checkpoint commits relative to the input branch.

If any gate fails or actionable or unsupported fixable vulnerabilities remain, do not push or open a PR. Preserve the local branch, checkpoint commits, current CVE state, and a precise failure summary.

Pull Request

When source changes exist and every final gate passes:

  1. Push the CVE-fix branch to origin.
  2. Open a ready-for-review PR in kubernetes-sigs/cloud-provider-azure targeting the input branch.
  3. Treat .github/PULL_REQUEST_TEMPLATE.md in the target checkout as authoritative. Start from its exact contents and preserve every section heading, the issue field, the fenced release-note and docs blocks, and any additional sections it contains. HTML guidance comments may be removed only after following their instructions. Never replace or drop repository template structure to match the skill asset; stop if the two cannot be merged without losing required content.
  4. Use title chore: fix cves for <input-branch>. Read assets/pull-request-template.md from the run-scoped snapshot of this skill, and replace every placeholder. Under What this PR does / why we need it, use Remediates actionable fixable vulnerabilities in the Linux CCM, CNM, and health-probe-proxy images for <input-branch>., followed by the populated results table and validation block from the asset. Under Special notes for your reviewer, use only the populated residual-risk details block from the asset. Do not use the asset as the complete PR body or copy its content into any other section.
  5. Preserve the repository template's heading text and order. Before opening the PR, verify that every #### heading from the repository template appears exactly once in the same order and that its release-note and docs fences remain. Use /kind cleanup, no issue closure unless the caller supplied one, and NONE in the release-note block. Replace the issue placeholder with NONE when no issue was supplied. Count vulnerability findings rather than grouped remediation actions in the baseline column. Use None instead of omitting an empty fix or residual-risk entry. For a shared root-module fix, identify the checkpoint that covers both CCM and CNM.

Cleanup and Output

In a final cleanup step on both success and failure, use the selected runtime to remove only the three recorded canonical references for this run's uniquely tagged local verification images. Ignore an image that was never created. Never run a system prune, remove shared base layers, clear Buildx cache, or delete images belonging to another run.

Clean fix-image-cves state after complete success. Preserve it after failure. On the no-change path, clean the state and remove the unused local CVE-fix branch.

Output:

ImageBaseline findingsApplied fixesFinal verificationResidual risks

Then report the module-consistency result, unit-test result, checkpoint commits, cleanup result, and PR URL, or the precise reason no PR was created.

© kubernetes-sigs, Apache-2.0. 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 1 other file (assets) in .agents/skills/remediate-image-cves of kubernetes-sigs/cloud-provider-azure.

  • SKILL.md
  • assets/pull-request-template.md

Open the folder on GitHubat commit 0201852

Compare with similar skills

Remediate Image Cves 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.

Remediate Image Cves compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Remediate Image Cves this skillkubernetes-sigs/cloud-provider-azure294—~3.9kAutomated safety check: PassApache-2.0
Performing Container Security Scanning With Trivymukul975/Anthropic-Cybersecurity-Skills34k—~818Automated safety check: PassApache-2.0
Defender For Containersvinayaklatthe/microsoft-security-skills175—~2.1kAutomated safety check: PassMIT
Container Securityhardw00t/ai-security-arsenal105—~2.8kAutomated safety check: PassNone
Sca TrivyAgentSecOps/SecOpsAgentKit2202 repos~3.7kAutomated safety check: PassCustom licence
Defender For Endpointvinayaklatthe/microsoft-security-skills175—~2.3kAutomated safety check: PassMIT

Similar skills

  • Performing Container Security Scanning With Trivy

    mukul975/Anthropic-Cybersecurity-Skills

    Runs Trivy across every target type it supports - container images, filesystems, Git repositories, and Kubernetes clusters - for OS and dependency vulnerabilities, IaC misconfiguration, exposed…

    34k GitHub stars~818 tokensUpdated 1 mo ago
    DevOps & CloudAuto-check passed
  • Defender For Containers

    vinayaklatthe/microsoft-security-skills

    Guidance for Microsoft Defender for Containers — Kubernetes and container security across AKS, Azure Arc-enabled Kubernetes, EKS, GKE, and OpenShift.

    175 GitHub stars~2.1k tokensUpdated 3 mo ago
    DevOps & CloudAuto-check passed
  • Container Security

    hardw00t/ai-security-arsenal

    Container and Kubernetes security assessment — image vulnerability scanning, SBOM diff analysis, K8s cluster auditing, RBAC privilege mapping, NetworkPolicy review, container escape testing, and…

    105 GitHub stars~2.8k tokensUpdated 5 mo ago
    SecurityAuto-check passed
  • Sca Trivy

    AgentSecOps/SecOpsAgentKit

    Software Composition Analysis (SCA) and container vulnerability scanning using Aqua Trivy for identifying CVE vulnerabilities in dependencies, container images, IaC misconfigurations, and license…

    220 GitHub starsUsed in 2 repos~3.7k tokens
    SecurityAuto-check passed
  • Defender For Endpoint

    vinayaklatthe/microsoft-security-skills

    Guidance for Microsoft Defender for Endpoint (MDE) — enterprise endpoint security with next-gen AV, EDR, attack surface reduction (ASR), Defender Vulnerability Management, automated investigation…

    175 GitHub stars~2.3k tokensUpdated 3 mo ago
    SecurityAuto-check passed
  • DefectDojo Vulnerability Management

    AgentSecOps/SecOpsAgentKit

    Aggregates scanner results into DefectDojo, deduplicates findings, tracks remediation SLAs and prepares compliance reports across products and pipelines.

    220 GitHub stars~2.3k tokensUpdated 5 mo ago
    SecurityAuto-check passed

More from kubernetes-sigs/cloud-provider-azure

All 12 skills in this repo
  • Run E2E Test

    kubernetes-sigs/cloud-provider-azure

    Official

    Parse a Go e2e test from tests/e2e/, translate each step to kubectl and az CLI commands, and interactively replay the test against a live cluster.

    294 GitHub stars~3.8k tokensUpdated yesterday
    Auto-check passed
  • Build Images

    kubernetes-sigs/cloud-provider-azure

    Official

    Build cloud-provider-azure container images through the repo Makefile with explicit IMAGETAG and IMAGEREGISTRY inputs, optional make flag overrides, and opt-in bounded Docker or Podman retries.

    294 GitHub stars~1.7k tokensUpdated yesterday
    Auto-check passed
  • Cherry Pick PR

    kubernetes-sigs/cloud-provider-azure

    Official

    Cherry-pick a merged pull request onto a release branch with Prow-style branch naming, manual conflict resolution, targeted validation, and GitHub PR creation.

    294 GitHub stars~636 tokensUpdated yesterday
    Auto-check passed
  • Create Release Note Doc PR

    kubernetes-sigs/cloud-provider-azure

    Official

    Generate or update the documentation-site release note for a given tag, commit it on a branch, push it to a writable remote, and open a GitHub PR to the docs branch.

    294 GitHub stars~768 tokensUpdated yesterday
    Auto-check passed
  • Create Release Tags

    kubernetes-sigs/cloud-provider-azure

    Official

    Create and optionally push the next Kubernetes-style release tag (vX.Y.Z) from a release-X.Y branch by resolving the remote branch tip, computing the next patch tag, and tagging the commit directly…

    294 GitHub stars~574 tokensUpdated yesterday
    Auto-check passed
  • Cve Remediator V2

    kubernetes-sigs/cloud-provider-azure

    Official

    Raise Go modules to caller-supplied minimum fixed versions from CVE/GO findings in any format, per tracked module root, sync go.mod/go.sum and root vendor/, audit the source module graphs, run the…

    294 GitHub stars~2.5k tokensUpdated yesterday
    Auto-check passed

Categories

Questions about Remediate Image Cves

What does Remediate Image Cves do?

Orchestrate end-to-end CVE remediation for the Linux CCM, CNM, and health-probe-proxy images on cloud-provider-azure master or a release-X.Y branch, including builds, repeated Trivy verification…. Remediate Image Cves is an agent skill from kubernetes-sigs/cloud-provider-azure, published by the product's own GitHub organization.Y branch, including builds, repeated Trivy verification, checkpoint commits, cleanup, validation, push, and pull-request creation.

When should I use Remediate Image Cves?

Remediate Image Cves fits situations like: the user wants to verify; fix CVEs on master; A release branch; run the three-image CVE workflow.

How do I install Remediate Image Cves in Claude Code?

Run `npx skills add kubernetes-sigs/cloud-provider-azure --skill remediate-image-cves -a claude-code`. Or copy the skill folder (.agents/skills/remediate-image-cves in kubernetes-sigs/cloud-provider-azure) into .claude/skills/remediate-image-cves in your project. Claude Code loads it when a task matches its description.

How do I install Remediate Image Cves in Codex?

Run `npx skills add kubernetes-sigs/cloud-provider-azure --skill remediate-image-cves -a codex`. Or copy the skill folder (.agents/skills/remediate-image-cves in kubernetes-sigs/cloud-provider-azure) into .agents/skills/remediate-image-cves in your project. Codex loads it when a task matches its description.

Can I use Remediate Image Cves 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 kubernetes-sigs/cloud-provider-azure --skill remediate-image-cves -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/remediate-image-cves, .gemini/skills/remediate-image-cves, .github/skills/remediate-image-cves and .opencode/skills/remediate-image-cves in your project.

What does Remediate Image Cves need to run?

Going by SKILL.md and its folder, Remediate Image Cves needs the command-line tools its instructions call (git, go, make and kind). Our summary lists: Docker.

Does Remediate Image Cves access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Remediate Image Cves 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. Review the folder before installing.

What licence does Remediate Image Cves use?

Remediate Image Cves is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Remediate Image Cves use?

About 3.9k tokens (SKILL.md is roughly 16k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Remediate Image Cves?

Skills that share tags, products or a category with Remediate Image Cves: Performing Container Security Scanning With Trivy (mukul975/Anthropic-Cybersecurity-Skills, 34k stars), Defender For Containers (vinayaklatthe/microsoft-security-skills, 175 stars), Container Security (hardw00t/ai-security-arsenal, 105 stars) and Sca Trivy (AgentSecOps/SecOpsAgentKit, 220 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Remediate Image Cves?

kubernetes-sigs (a GitHub organization, an official publisher) maintains it in kubernetes-sigs/cloud-provider-azure, which has 294 GitHub stars. The repository holds 12 skills in this directory. The repository was last updated on October 9, 2026.

Source: kubernetes-sigs/cloud-provider-azure on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.