AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

15 Factor Observability

skill-just-agent-just-product-factor-15-factor-observability · by Just-Agent

A Claude skill from Just-Agent/just-product-factor.

No reviews yet
0 installs
16 views
0.0% view→install

Install

$ agentstack add skill-just-agent-just-product-factor-15-factor-observability

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-just-agent-just-product-factor-15-factor-observability)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of 15 Factor Observability? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

15-factor-observability Skill

Skill 名称

15-factor-observability

适用场景

Use this Skill when a product, Agent, app, RAG system, MCP server, CLI, workflow, or service needs clearer runtime visibility, logs, metrics, traces, audit artifacts, failure diagnostics, and operator handoff.

适用于这些情况:

  • 项目能运行,但失败后不知道发生了什么。
  • Agent 有工具调用、长任务、异步任务、RAG 检索、MCP 调用或 CI/CD 工作流,但缺少可追踪证据。
  • README 写了功能,但没有说明如何观察运行状态、定位错误、复现问题。
  • 开源项目要发布,需要基础日志、调试信息、验证产物和 issue 诊断路径。
  • 需要把一次 Agent 优化过程变成可复盘的审查记录。

不适用场景

  • 只需要概念解释,不需要审查项目。
  • 项目仍处于纯草稿阶段,连最小运行路径都不存在。
  • 用户只关心视觉样式或文案,此时优先使用 10-factor-html6-factor-docs

输入要求

优先读取:

  • README、quickstart、docs、examples
  • 源码中的日志、错误处理、运行入口、配置文件
  • CI/CD workflow、测试脚本、release 脚本
  • Agent trace、tool call 记录、RAG 检索日志、MCP tool schema
  • issue 模板、bug report 模板、validation summary、audit report
  • 运行失败日志或用户报告的问题

工作流程

  1. 明确系统的关键运行路径和用户承诺。
  2. 找出失败点:安装、配置、命令、网络、权限、工具调用、数据输入、部署、发布。
  3. 检查日志是否结构化、可定位、可脱敏、可复现。
  4. 检查 metrics / traces / run artifacts 是否覆盖关键路径。
  5. 检查 Agent、RAG、MCP、CI/CD 是否留下足够审计证据。
  6. 使用 checklist.md 做逐项审查。
  7. 使用 scoring-rubric.md 评分。
  8. 使用 audit-report-template.md 输出审查报告。
  9. 使用 refactor-plan-template.md 输出分阶段改造计划。
  10. 如涉及安全、隐私或敏感日志,联动 14-factor-security-privacy

审查维度

  1. Runtime map:关键运行路径是否清楚。
  2. Structured logging:日志是否结构化、分级、可搜索。
  3. Error diagnostics:错误是否可行动、可复现。
  4. Metrics:是否有基础计数、耗时、失败率和吞吐指标。
  5. Tracing:是否能串起请求、Agent step、tool call、workflow job。
  6. Artifacts:是否保存审查报告、运行摘要、验证结果、失败样例。
  7. Privacy-safe observability:日志是否避免泄露 secrets、PII、tokens、prompt 原文。
  8. Debug UX:用户和维护者是否知道如何提交有用 bug report。
  9. CI/release visibility:自动化流程是否产出可读验证结果。
  10. Agent handoff visibility:Agent 是否能基于日志和产物继续迭代。

评分方式

使用 0-5 分:

  • 0:完全缺失。
  • 1:有零散日志,但不可复现。
  • 2:有基本日志和错误信息,但缺少结构和关键路径覆盖。
  • 3:覆盖主要路径,有基础验证产物。
  • 4:日志、指标、trace、产物、issue 诊断路径基本完整。
  • 5:可发布级 observability,支持用户、维护者和 Agent 持续定位问题。

重构优先级

优先级顺序:

  1. 先让失败可见:安装、运行、测试、构建、核心命令。
  2. 再让失败可定位:错误码、上下文、建议修复动作。
  3. 再让运行可复盘:audit summary、validation summary、decision log。
  4. 再让系统可度量:耗时、次数、失败率、覆盖率。
  5. 最后接入更高级 tracing / dashboards / release reporting。

输出格式

输出必须包含:

  • Observability diagnosis
  • 15-factor scorecard
  • Critical blind spots
  • Recommended logs / metrics / traces / artifacts
  • Privacy-safe logging notes
  • Refactor plan
  • Validation plan
  • Paired Skill recommendations

示例调用

Use the 15-factor-observability skill to audit this Agent project for logs, traces, tool-call evidence, validation artifacts, and debuggability.
Use the 15-factor-observability skill with 14-factor-security-privacy to make this app easier to debug without leaking secrets or user data.

版本记录

v0.9.0:首次加入 observability 标准,用于补齐运行可见性、失败诊断、审查产物和 Agent handoff 证据。

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.