everruns/plugins/everruns-operate

Claude Code and agent skill by everruns/plugins

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Answer organization-level questions about Everruns. How much was spent and on what, token and tool usage by agent or model, what is broken (health issues, failing turns), who changed what (change history, audit log), and budgets. Use when the user asks about usage, cost, reliability or recent changes across agents rather than about one session. Builds on the everruns skill.

Install

npx vskill@latest install everruns/plugins/everruns-operate

SKILL.md

Operate an Everruns organization

Everything here is read-only unless the user asks for a change, so run it with MCP query or the CLI. Commands are the everruns command language (see the everruns-cli skill); check flags with --help.

Start broad with a report, then drill into the agents and sessions it points at.

Usage and spend: reports

bash
everruns reports catalog get # datasets, dimensions, measures, filters
everruns reports query run --help # one query: dataset, measures, dimensions, filters, time range

The catalog is the source of truth; read it before writing a query. The datasets today:

DatasetAnswers
llm_generationsTokens, latency, time to first token, errors by model, provider, agent or day
tool_callsWhich tools run, how often, and how often they fail
turns, sessionsVolume and outcomes of agent work
budget_postingsMoney: debits and credits against budgets
capability_usageWhich capabilities are actually used

Group by agent_name or model and by day for trends. Saved reports (everruns reports saved list) are ones the organization already relies on: check them before building a new query.

Report numbers with their time range and units, and say which dataset they came from.

Budgets

bash
everruns budgets list --help # limits per session, agent, user or organization
everruns budgets ledger list --help # what was charged and when

Creating a budget, changing a limit or topping up spends or restricts money: confirm the subject, amount and currency with the user first, then use execute with --reason.

What is broken

  1. everruns health-issues list: problems Everruns already detected (provider

credentials, failing integrations). everruns health-issues get <id> explains one.

  1. Failing work: query turns or llm_generations grouped by agent_name for

errors over the last day, then open a failing session (everruns-debug-session).

  1. Per-agent checks: everruns agents health-checks list --help.

Who changed what

  • everruns history org --help: every recorded change across the

organization, newest first, with the reason given.

  • everruns history list <id> and everruns history diff --help for one entity.
  • everruns orgs audit-logs list --help: sign-ins, membership and other

security events.

When a behavior change lines up with a recorded change, say so with both timestamps. Rolling back (history restore) is a change: ask first.

Models and providers

  • everruns models list, and the organization default model.
  • everruns providers list --help: which model providers are configured.

Provider keys are entered by the user on the Providers page, never through you.

Reporting back

Lead with the answer (a number, a list of broken things, the change that explains it), then the evidence: the query or command, the time range, and links to the agents and sessions involved.

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/everruns/plugins/everruns-operate/badge)](https://verified-skill.com/skills/everruns/plugins/everruns-operate)

Something wrong with this skill or its page? Tell us.