Investigate production issues by correlating logs, metrics, traces, and dashboards.
npx skills add https://github.com/samcdavid/dotfiles --skill my-investigate
Please help me install this Agent Skill. Skill: my-investigate Source: https://github.com/samcdavid/dotfiles/tree/main/claude/skills/my-investigate Command: npx skills add https://github.com/samcdavid/dotfiles --skill my-investigate
Investigate production or runtime issues by exploring logs, metrics, traces, and dashboards. Product-agnostic — works with any observability stack. Follows the evidence to root cause.
Describe the issue symptoms and gather available logs, metrics, and traces from the observed time window to start the investigation.
High-intent search queries and answers about installing and using this skill.
To investigate production incidents, you collect time-aligned logs, metrics, and traces across affected services. This process builds an evidence-based timeline to systematically test hypotheses and pinpoint the root cause of the runtime issue.
Finding a root cause across services requires a platform-agnostic approach to correlate observability data. By applying a structured hypothesis-testing workflow to logs and traces, you can follow the evidence end-to-end from event to resolution.
Yes, this incident investigation workflow is platform-agnostic and works with any observability stack. It requires no specific dependencies, allowing you to gather data from any available logs, metrics, traces, and dashboards.
To start an evidence-driven investigation, describe the issue symptoms and gather available logs, metrics, and traces from the observed time window. This initial data collection forms the baseline for your root-cause analysis.
You need a structured hypothesis-testing workflow when a runtime issue requires end-to-end evidence-based analysis across logs, metrics, and traces. It guides the correlation of time-aligned data to methodically identify root causes rather than guessing.
lygavInstrument federated squad agents with OpenTelemetry to collect traces, metrics, and logs.
leonardoteodorooGenerate blameless postmortems with root cause analysis, timelines, and action items.
weirdgmeInstrument distributed systems with OpenTelemetry and correlate logs, metrics, and traces.
FrekiManagarmConfigure PostHog exception tracking for Node.js applications with source maps.
