- Home
- Skills
- DevOps & Cloud
- opspilot infrastructure diagnostics
Works with the AI Agents you already use
opspilot infrastructure diagnostics
Diagnostic triage for PostgreSQL, Docker, and Kubernetes using the OpsPilot MCP server.
$17
opspilot infrastructure diagnostics
Example session with this skill installed
The 'api-gateway' pod in the 'production' namespace is crashing. Use OpsPilot to find out why and check if it is related to our Postgres database.
- Read your context and instructions
- Compiled the opspilot infrastructure diagnostics
I have investigated the 'api-gateway' pod in the 'production' namespace and checked for correlations with the PostgreSQL backend using OpsPilot diagnostics.
Infrastructure Diagnostic Report: api-gateway
| Category | Resource | Status | Findings |
|---|---|---|---|
| Kubernetes Pod | api-gateway | Failed | Pod status is CrashLoopBackOff. Last termination reason: Error (Exit Code 1). |
| K8s Events | production | Warning | Back-off restarting failed container gateway-main. |
| Pod Logs | api-gateway | Log Trace | FATAL: password authentication failed for user "gateway_user" |
| PostgreSQL | db-primary | Healthy | Database is reachable. postgres_health returned OK. |
| Correlation | Cross-System | Identified | diagnose_cross_system_issues confirms the pod failure is due to rejected DB credentials. |
Diagnostic Evidence
-
Observation: The pod logs explicitly show a
FATALauthentication error during the application startup sequence.
Hypothesis: The Kubernetes Secret containing the database password may be out of sync with the PostgreSQL user roles, or the gateway_user credentials have been rotated without updating the deployment.
Postgres Status: The database itself is accepting connections, confirming this is an authentication/config issue rather than a database outage.
Next steps
- Verify the
POSTGRES_PASSWORDvalue in the Kubernetes Secret for theproductionnamespace. - Run
postgres_list_tablesto confirm the user has appropriate schema permissions once authenticated. - Check for recent CI/CD jobs that might have rotated database credentials.
Connects securely to your tools. The creator never sees your data.
About this skill
Infrastructure failures in PostgreSQL, Docker, or Kubernetes often lead to hours of manual log digging and command-line context switching. This skill integrates with the OpsPilot MCP server to provide evidence-based triage and diagnostics directly within your agent workflow.
What it does
- Database health checks Executes
postgres_healthand table listing to identify connectivity or schema-related access issues. - Container inspection Retrieves container states, health status, and metadata via
docker_inspect_containeranddiagnose_docker_issues. - Kubernetes pod triage Lists namespaces, pods, and events to isolate failures in specific clusters or deployments.
- Log retrieval Pulls targeted logs from Docker or Kubernetes pods with configurable tail limits to avoid noise.
- Cross-system correlation Uses
diagnose_cross_system_issuesto identify failures spanning both Kubernetes pods and PostgreSQL backends.
How it works
- Verify connectivity to the local OpsPilot MCP server and identify authorized backend targets.
- Select the specific troubleshooting workflow for the resource type (DB, Container, or Pod).
- Execute diagnostic commands to gather evidence, events, and logs.
- Synthesize tool output into a report distinguishing between confirmed observations and hypotheses.
Frameworks & tools
This skill requires a separate local installation of the OpsPilot MCP server. It interfaces with PostgreSQL databases, Docker engine via local sockets, and Kubernetes clusters using standard kubeconfig contexts.
Why this beats prompting it yourself
Writing manual prompts for infrastructure debugging often results in generic advice or dangerous remediation suggestions. This skill enforces a strict observational workflow, uses specialized diagnostic tools instead of raw shell commands, and handles the nuances of cross-system correlation between Kubernetes and Postgres.
Use cases
- Debugging a
CrashLoopBackOffin a Kubernetes pod to find the root cause. - Investigating intermittent PostgreSQL connection failures from an application.
- Auditing Docker container health when services are unresponsive but running.
- Correlating pod restarts with database backend latency or schema locks.
Known limitations
Requires a pre-installed OpsPilot MCP server. It provides observations and recommendations only and cannot perform autonomous remediation or infrastructure changes.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for DevOps & Cloud.
Requires separate OpsPilot MCP server installation. Observation only; no remediation capabilities. Docker is excluded from cross-system correlation. Log tailing limited to 1000 lines.
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 buying a skill.
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
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 5 days ago
- Accounts
Postgresql
Docker
- Software
- Opspilot MCP
Python 3.10+
GitAI coding agent with MCP stdio support
Access to PostgreSQL, Docker,orKubernetes
Install OpsPilot MCP separately and configure only the backends you need. PostgreSQL requires credentials, Docker requires daemon access, and Kubernetes requires appropriate cluster permissions. Use restricted access in trusted environments.
Permissions
Allowed Hosts
File Scopes
OpsPilot uses local MCP server execution to perform read-oriented infrastructure diagnostics. Terminal and file access support installation and MCP configuration. Network access and environment variables support connections to user-configured PostgreSQL and Kubernetes systems. Docker diagnostics require daemon access, which is highly privileged. Credentials should be restricted, and diagnostic data may contain sensitive information. OpsPilot does not automatically modify or repair infrastructure.