agentleFS
rubixkube-iorubixkube-io/rubixkube-for-ai/skills/knowledge/SKILL.md

knowledge

Skill1 starsChanged 35 days agoMITView on GitHub

What RubixKube has recorded about the organization's systems and history: the knowledge graph of resources, incidents, RCAs, changes and people. Use when the person asks about a service's history, who worked on something, what fails most, or any question whose answer is in past incidents rather than live state.

13 other projects use a version of this.

Adopted by

Most recently taken up by

The file

Files
SKILL.md

When to use

  • "What's the history of the payments service?" / "Has api-gateway failed before?"
  • "Who has dealt with the checkout incidents?" / "Which resources fail most?"
  • "What do we know about node pool X?"

Tools

  • search_knowledge(phrase, entity_type?, hops?, limit?): find a service, host, resource or person by name and see what it connects to. Each hit is the matched node, the nodes around it and the relations between them as sentences (Incident:… AFFECTS Resource:…). entity_type narrows to one kind: Service, Resource, Incident, RCA, Change, User.
  • ask_knowledge(question): one plain-words question, one short answer from the recorded history. Use it for questions across many things; use search_knowledge for one thing.
  • From an incident, prefer similar_incidents and what_changed (see the investigate skill); they already know the incident's resources and fingerprint.

How to answer

Say what the graph recorded. It holds incidents, RCAs, Tasks, changes, environments and people as they happened, not metrics or logs, and not the state right now. For the current state of a source call environment_snapshot; for one incident call get_incident. If the graph has nothing, say so and offer list_incidents. Do not add to the graph from here; that is done by RubixKube itself as incidents and RCAs happen.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.