Skip to main content Skills Marketplace コミュニティが作成したAIスキルを発見・探索
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/domehahn/skil --skill observability-reviewerコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Zipをダウンロード ダウンロード中... name observability-reviewer description Describe what this skill helps an agent do. version 0.1.0 since 2026-07-28 last_modified 2026-07-28 authors ["platform-engineering"] stability experimental min_platform_version {"codex":"unknown","amazon-q":"unknown","antigravity":"unknown","auggie":"unknown","bob":"unknown","claude-code":"unknown","cline":"unknown","codebuddy":"unknown","continue":"unknown","costrict":"unknown","crush":"unknown","github-copilot":"unknown","gitlab-duo":"unknown","factory":"unknown","forgecode":"unknown","opencode":"unknown","openhands":"unknown","cursor":"unknown","roo-code":"unknown","kiro":"unknown","junie":"unknown","gemini-cli":"unknown","iflow":"unknown","kilocode":"unknown","kimi":"unknown","lingma":"unknown","pi":"unknown","qoder":"unknown","qwen":"unknown","windsurf":"unknown","ollama":"unknown"} deprecated_since null replaces null supersedes [] changelog [{"version":"0.1.0","date":"2026-07-28","change":"Initial generated production-ready SDLC / DevSecOps skill"}]
Observability Reviewer
Purpose
Review logging, metrics, tracing, alerting, dashboards, operational visibility, SLOs, SLIs, runbooks, audit logs, sensitive data in logs, on-call usability, and incident detection.
When to use
A service, feature, or deployment changes behavior that operators must detect, debug, or support.
Logs, metrics, traces, alerts, dashboards, or runbooks are added or changed.
SLOs, SLIs, audit logs, or incident detection coverage is unclear.
Sensitive data may enter logs or telemetry.
The central agent routes to observability review.
Operating model
Map critical user journeys, failure modes, dependencies, and operational questions.
Check whether logs, metrics, traces, alerts, and dashboards answer those questions.
Review signal quality: labels, cardinality, correlation IDs, thresholds, routing, and ownership.
Assess privacy and security of telemetry.
Recommend concrete telemetry, alert, dashboard, or runbook changes.
Spec-Driven Change Context
Treat repository specs, ADRs, runbooks, change proposals, design notes, and task files as durable context that outlives a chat session.
For non-trivial changes, prefer a checked-in change artifact or equivalent proposal/design/tasks record before implementation begins.
Capture requirement deltas explicitly: added, modified, removed, deprecated, or unchanged behavior.
Keep implementation tasks traceable to acceptance criteria, affected specs, validation commands, and owners.
During verification, compare the implementation against the proposal, design decisions, task checklist, and spec deltas.
After completion, sync or archive completed change artifacts so the repository's source of truth reflects the final behavior.
If the repository has no spec workflow yet, report the missing artifact and provide a minimal proposal/spec/tasks outline instead of relying on chat-only intent.
Skill-Specific Review Scope
Logs, metrics, traces, alerts, dashboards, and runbooks.
SLOs, SLIs, critical journeys, audit logs, and incident detection.
Sensitive data in logs, correlation IDs, retention, and cardinality.
On-call usability, alert routing, alert fatigue, and ownership.
Operational decision support and recovery verification.
Skill-Specific Checklist
Decision Rules
If operators cannot detect failure of a critical journey, require metrics or alerts before release.
If logs can expose secrets or PII, require redaction before approval.
If an alert lacks owner, severity, runbook, or action, classify as alert-quality gap.
If dashboards cannot support rollback decisions, require deployment health panels.
If metric cardinality can explode from user input, require label redesign.
If audit-relevant actions lack logs, raise security/compliance severity.
Finding Categories
Missing signal for critical user journey or dependency failure.
Unsafe logging of secrets, PII, tokens, or sensitive payloads.
Unactionable, noisy, duplicate, or ownerless alert.
Dashboard gap for deploy health, customer impact, or rollback decision.
Trace, correlation ID, or context propagation gap.
SLO/SLI, audit log, retention, or cardinality risk.
Severity Guidance
Critical: immediate exploitability or operational failure can expose secrets, regulated data, production safety, or release integrity.
High: credible security, reliability, compliance, rollback, or user-impact risk requires owner action before merge or release.
Medium: meaningful maintainability, validation, documentation, or process gap should be tracked and resolved.
Low: advisory improvement, clarity issue, or hardening opportunity with limited immediate impact.
DevSecOps Guardrails
Do not read secrets, .env files, private keys, production credentials, masked CI/CD variables, database dumps, or sensitive logs unless explicitly required.
Do not push, deploy, publish, merge, or create releases unless explicitly asked.
Prefer merge requests, reviewable diffs, and auditable validation evidence.
Prefer least privilege, minimal changes, and explicit rollback notes.
Do not fabricate test results, repository state, commands, security findings, or validation outcomes.
Report assumptions, uncertainty, residual risk, and validation gaps clearly.
Output Requirements
Observability coverage map for journeys, failure modes, signals, alerts, dashboards, and runbooks.
Findings with affected signal, evidence, operator impact, and remediation.
Concrete metric, log, trace, alert, dashboard, or runbook recommendation.
Sensitive telemetry and cardinality risk assessment.
Post-deploy monitoring and incident-detection recommendation.
Residual blind spots and owner actions.
Acceptance Criteria
Critical journeys have metrics, logs/traces, alerts, and dashboard support.
Telemetry excludes secrets and sensitive payloads.
Alerts are actionable, owned, routed, and tied to runbooks.
Dashboards support diagnosis and rollback decisions.
SLOs/SLIs or audit logs exist where required.
Known blind spots are explicitly documented.
Anti-Patterns
Adding logs without deciding what operator question they answer.
Alerting on every error without severity, owner, or action.
Using high-cardinality user input as metric labels.
Logging full payloads to debug production issues.
Building dashboards that cannot support rollback or incident triage.
Assuming tracing solves missing metrics or alerting.
Changelog
0.1.0 - 2026-07-28
Initial generated production-ready SDLC / DevSecOps skill.
