Normalize agentic execution into runs, events, and commands for operator dashboards.
npx skills add https://github.com/thistleknot/skills --skill build-observability
Please help me install this Agent Skill. Skill: build-observability Source: https://github.com/thistleknot/skills/tree/main/build-observability Command: npx skills add https://github.com/thistleknot/skills --skill build-observability
Build observability for agentic systems is often missing at the exact moment operators need it most, leaving you with only raw logs and no reliable way to answer what is running, what changed, what artifacts exist, and how the run ended.
Use the build-observability skill to instrument your agent runtime by projecting its session/transcript artifacts into the normalized runs/events/commands schema and wiring the dashboard to those projections.
High-intent search queries and answers about installing and using this skill.
Build observability for agentic coding by projecting runtime session artifacts into normalized runs, events, and commands tables. This creates stable, query-friendly schemas that power operator-ready dashboards showing current stage, timeline, and command lifecycle.
Trace enrichment for agent observability adds decision context to normalized execution events. You need it when transforming raw run tracking data into episodic inputs for memory and context-graph systems, ensuring observability captures why decisions were made, not just what executed.
Yes, this observability approach defaults to SQLite for local harnesses and supports optional Postgres for shared or remote readers. The storage policy enables local-first run tracking with seamless escalation to shared dashboard access when multiple operators need visibility.
Normalize agent runtime logs by applying a collector adapter pattern that maps local session transcripts, files, and metadata into a canonical runs, events, and commands schema. This avoids hard-coding foreign layouts while projecting artifacts into query-friendly dashboard tables.
Track file changes and command lifecycles by projecting multi-stage agent execution into normalized tables that capture command activity, changed files, artifacts, and deploy outcomes. This provides operator-focused dashboard visibility without parsing raw logs manually.
Yes, a runtime collector adapter is required to adapt your specific agent runtime artifacts into the normalized schema. It bridges local session formats and the canonical runs, events, and commands projections needed for operator-ready dashboard visibility.
starrank-softParse and sanitize agent JSONL logs into Markdown transcripts for GitHub PRs.
metratonSearch and retrieve episodic memory entries from Gaia's Memory v2 index.
marcioaltoeDesign normalized PostgreSQL schemas and Drizzle ORM definitions.
ppiankovCaptures, triages, and analyzes ephemeral Kubernetes logs in real-time via CLI and sidecar injection.
