Skip to main content

secure-an-internal-api

Secure an internal API with workload identity, authorization, transport protection, input limits, audit, segmentation, and abuse resistance. Use when a service boundary is private but still consequential.

来源信息

仓库
protosphinx/sphinxstack
最近来源活动
2026年7月29日 13:55
检测到的 SKILL.md 语言
英语
星标
3
Forks
0

安装方式

默认使用会先检查来源的 Prompt;你也可以切换为直接命令,或下载本地副本。

检查来源文件

决定是否安装前,请先阅读 SKILL.md,以及 SkillsMP 当前展示的配套文件。

文件资源管理器
2 个文件

正在显示 SKILL.md

SKILL.md
来源说明 · 只读预览
name
secure-an-internal-api
category
code
description
Secure an internal API with workload identity, authorization, transport protection, input limits, audit, segmentation, and abuse resistance. Use when a service boundary is private but still consequential.
# secure-an-internal-api ## When to use - Use for service, admin, operations, or partner-only APIs. - Never treat network location alone as identity or authorization. ## Preconditions - Inventory callers, actions, resources, data, networks, identities, failure modes, and emergency access. ## Procedure 1. Assign authenticated workload identities with short-lived credentials. 2. Authorize each action and resource with tenant and purpose context. 3. Protect transport and verify peer identity end to end. 4. Validate schemas, sizes, rates, deadlines, replay, and idempotency. 5. Segment sensitive routes and restrict outbound dependencies. 6. Audit consequential decisions without logging secrets or payloads. 7. Test confused deputy, credential theft, SSRF, replay, overreach, and outage. ## Failure plan - Fail closed for consequential action and preserve a bounded governed recovery path. ## Worked example A private billing endpoint rejects a valid service identity that lacks authority for the requested tenant. ## Done - An internal-API security report records identities, authorization, transport, validation, segmentation, audit, and recovery - Cross-tenant, replay, SSRF, rate, credential, failover, log-redaction, and emergency-access tests verify controls
在 GitHub 查看