Skip to content

lidge-jun/codexclaw / plugins/codexclaw/skills/dev-devops

cxc-dev-devops

Verified

Use for infra, CI/CD, releases, repo policy, worktree cleanup and native desktop apps. Triggers: Dockerfile, K8s, IaC, SRE, Tauri, 스택 PR CI, 배포, 인프라, 쿠버네티스, 브랜치 정리, 브랜치 삭제, 워크트리 정리, 저장소 세팅, 브랜치 보호, 에이전트 PR, PR 정책, 데스크톱 앱, 메뉴 막대.

references/agent-infra-safety.md

View on GitHub
# Agent Infrastructure Safety (DEVOPS-AGENT-SAFETY-01, DEFAULT)

Source: sol research (microsoft/skills governance, cloudflare/skills sandbox, google/skills).

## Execution Model
- Default mode: PLAN-ONLY. The agent generates the plan; a human approves execution.
- Write mode: explicitly granted per task. Scoped to declared resources only.
- Destructive actions (delete, scale-to-zero, revoke, drop): require explicit confirmation even in write mode.

## Credential Separation
- Read credentials: service account with read-only IAM bindings
- Write credentials: separate service account with scoped write permissions
- Never share a single credential across read and write operations
- Prefer workload identity / OIDC over long-lived secrets

## Command Allowlist
- Agents should operate through declarative tools (Terraform plan, kubectl diff, helm template)
- Direct imperative commands (kubectl delete, aws ec2 terminate) require explicit approval
- Shell pipes and command chaining are suspicious in infrastructure context

## Audit Correlation
- Every infrastructure action is tagged with: agent session ID, task ID, operator identity
- Actions without correlation tags are audit failures
- Retain action logs for the organization's compliance window

## Sandbox and Egress
- Infrastructure agents run with network egress restricted to declared endpoints
- Cloud API endpoints are allowed; arbitrary internet access is not
- File system access limited to the working directory and declared output paths

## Approval Classes
| Action class | Approval required |
|-------------|-------------------|
| Read/inspect/plan | None (read credentials sufficient) |
| Create/update (non-destructive) | Task-level write grant |
| Delete/revoke/scale-down | Explicit per-resource confirmation |
| Cross-account/cross-region | Elevated approval + blast-radius statement |
| Production environment | Always requires human confirmation |

## Pull-request intake
This file governs agents operating infrastructure. Agents *contributing* pull
requests are governed by `agent-pr-intake.md` (DEVOPS-AGENT-INTAKE-01); the approval
classes above do not map onto PR review.
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 321kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 197kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall