CatalogueDelivery & Operations
Observability & Instrumentation Gap Audit
Finds the paths where a failure would be invisible.
What it examines
- Requests with no log line that identifies them
- Missing metrics on latency, errors, and saturation
- Traces that stop at the first process boundary
- Alerts that do not correspond to a user-visible failure
What you get back
A findings file. Each item has the evidence, how serious it is, and a fix.
Dependencies
- An AI coding agent that reads SKILL.md files, such as Claude Code, Codex or Gemini CLI
- Read access to the repository: application code, logging configuration and any tracing or metrics SDK setup
- Python 3.11 or newer with the jsonschema package, to check findings.json before fixing anything
- Optional: log aggregation and shipping configuration
- Optional: access to the metrics or tracing backend, or the vendor dashboard. Alerting that lives only in a vendor UI is not visible from a repository scan, so without access it is recorded as insufficient evidence, not as absent.
How it works
The audit is read-only. It runs seven checks: log structure quality, correlation id propagation, business metric coverage, alert actionability, health check semantics, SLO definition presence and log aggregation coverage.
Correlation ids are traced end to end on a real business-critical path, including asynchronous hops. An installed dependency is not treated as proof that something is instrumented.
It never invents SLO thresholds. findings.md carries a list of questions for you to answer instead.
The remediate skill reads findings.json only. It starts at service boundaries, prefers auto-instrumentation, migrates logging one module at a time and never adds personal data or secrets to spans or logs.
Setup
- Unzip the download, then copy the audit and remediate folders into your agent's skills folder (for Claude Code, .claude/skills/) in the repository you want to check.
- List your business-critical paths. The audit traces them.
- Ask your agent: Run an observability and instrumentation gap audit on [scope]. Business-critical paths: [list them]. Mode: audit only.
- Check the result with the validator: pip install jsonschema, then python3 schema/validate.py findings.json.
- Read findings.md. Once you have reviewed the findings, run the remediate skill using findings.json, or re-run with Mode: audit and remediate.
Run it with your agent
Copy the prompt, fill in the parts in square brackets, and paste it into your agent.
Run the e3-observability-instrumentation-audit skill on this repository. Business-critical paths: [list them]. Mode: audit only. Do not change any code. Write findings.json and findings.md in the repository root.
Run the e3-observability-instrumentation-remediate skill using findings.json. Work through the findings I choose, and leave the rest open.