- Home
- Skills
- DevOps & Cloud
- Vargr Log Format Advisor
Works with the AI Agents you already use
Vargr Log Format Advisor
Score your logging out of 100 and get the exact fixes that make incidents faster to debug.
Free
Vargr Log Format Advisor
Example session with this skill installed
Audit the logging in this repo. It's a FastAPI service that ships logs to Elasticsearch.
- Read your context and instructions
- Compiled the vargr log format
✅ Worked
Summary
Logging score: 58/100. Diagnosable with effort: logs are JSON but have no trace ids and no service version, and one endpoint leaks emails.
⚠️ Privacy (must fix)
• app/routes/auth.py:41 logs the full request body on login failures → emails and passwords reach Elasticsearch. Log user_id (opaque) and the reason instead.
Top 3 fixes by incident impact
- Add service.version to every line (from the image tag) → version correlation becomes one query.
- Add trace.id/span.id from OpenTelemetry → follow one request across services.
- Log exceptions once, in the exception handler, as error.type / error.message / error.stack_trace fields.
Findings
| Area | Where | Problem | Fix |
|---|---|---|---|
| Correlation | logging_config.py | no trace.id | ecs_fields processor |
| Errors | orders/service.py:88 | exception logged at 3 layers | log at the boundary only |
| Levels | middleware.py:20 | 404s logged as ERROR | WARN or INFO |
| Noise | health.py:12 | /health logged every 5 s (31% of volume) | drop or sample |
Proposed change: logging_config.py (structlog + ECS fields), shown as a diff.
Next steps
• Apply the diff? I can open a draft PR.
Connects securely to your tools. The creator never sees your data.
About this skill
Good logs turn a two-hour incident into a ten-minute one. This skill audits how your code writes logs, scores it out of 100 for diagnosability, and proposes concrete changes in your own language and logger.
It checks structure (JSON), ECS / OpenTelemetry field names, trace and span ids, service version, how exceptions are captured, event naming, log levels, noise, and privacy leaks (tokens, passwords, emails, card and ID numbers).
Includes ready-to-use logger configurations for Python (structlog / ecs-logging), Node.js (pino), Java (Spring Boot / Logback), .NET (Serilog) and Go (slog), plus a redaction script to check real log samples for leaks.
Part of Vargr: Vargr Tracer finds the root cause of incidents and opens the fix; Vargr Hunter spots attacks and anomalies.
Requires access to logger configuration files and source code call sites. Privacy scanning for sample logs requires the redact.py script included in the Vargr plugin root.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 26 days ago
- Free with an Agensi account
Permissions
• Read files in the current repository. • Optional: run scripts/redact.py (Python standard library, no network access) on log samples you provide. • Writes files only if you ask it to apply changes.