Plugin Claude Code
Claude Code plugin for Cekura AI voice agent testing and evaluation.
6 3d ago A
tokens not measured
original MIT
Instructions file
Instructions for cekura-ai/cekura-skills, covering cekura skills — developer guide, repository structure, two install paths, one source of truth, skill content rules and update workflow.
6 3d ago A 5,665 tokens
original MIT
Instructions file Gemini CLI
Instructions for cekura-ai/cekura-skills, covering cekura platform — agent guidelines, core concepts, api basics, part 1: metric design and the metric creation workflow.
6 3d ago A 8,589 tokens
original MIT
cekura
04Plugin Claude Code
Cekura voice AI testing & evaluation — platform onboarding, agent setup, metric design and improvement, evaluator design and execution.
6 3d ago A
tokens not measured
original MIT
cekura
05MCP server Claude CodeCodexCursor +2
MCP server "cekura", hosted remotely at api.cekura.ai, as configured in cekura-ai/cekura-skills.
6 3d ago A
tokens not measured
original MIT
Agent
Use this agent to design comprehensive eval test coverage from an agent description. This agent analyzes the agent's workflows, decision points, and edge cases to produce a complete test coverage plan. Examples: Context: The user has a new agent and needs to build a test suite from scratch. user: "I need to create…
6 3d ago A 320 tokens
original MIT
Agent
Use this agent to review Cekura metric quality and simulate the labs improvement workflow. This agent should be triggered proactively after creating or modifying metrics, or when the user wants to systematically improve metric accuracy. Examples: Context: The user just created a new metric using the create-metric…
6 3d ago A 302 tokens
original MIT
autogen-eval
08Command
Auto-generate Cekura evaluators using the generate API with full configuration.
6 3d ago A 20 tokens
original MIT
Command
State-aware Cekura onboarding that resumes at the right setup phase and delegates to the cekura-onboarding skill. Supports two variants — testing (default) and observability.
6 3d ago A 43 tokens
original MIT
Command
Run a full end-to-end agent quality report — generates 10–20 evals, runs them, and produces a structured markdown report.
6 3d ago A 31 tokens
original MIT
Command
Create or update a Cekura metric for evaluating voice AI agent calls.
6 3d ago B 18 tokens
original MIT
eval-results
12Command
Fetch Cekura eval results and summarize common workflow issues.
6 3d ago A 14 tokens
original MIT
Command
Run specific Cekura metrics against selected calls for evaluation or re-evaluation.
6 3d ago A 19 tokens
original MIT
Command
Improve a Cekura metric through feedback collection, labs pipeline, and auto-improvement.
6 3d ago A 22 tokens
original MIT
list-evals
15Command
List Cekura evaluators filtered by agent, project, or tags.
6 3d ago A 18 tokens
original MIT
list-metrics
16Command
List Cekura metrics filtered by agent or project.
6 3d ago A 14 tokens
original MIT
Command
Manually create, update, or duplicate a Cekura evaluator (a.k.a. scenario, eval).
6 3d ago A 28 tokens
original MIT
report-bug
18Command
Report a bug in a Cekura skill, command, or MCP integration.
6 3d ago B 19 tokens
original MIT
run-evals
19Command
Execute Cekura evaluators (voice, text, websocket, sip, pipecat, vapi, retell, elevenlabs, livekit).
6 3d ago A 35 tokens
original MIT
setup-mcp
20Command
Configure the Cekura MCP server for the cekura plugin.
6 3d ago B 16 tokens
original MIT
Command
Update Cekura skills to the latest version from GitHub.
6 3d ago A 16 tokens
original MIT
SessionStart
22Hook
Runs when a session starts, executing claude-self-update.sh. From cekura-ai/cekura-skills.
6 3d ago A
tokens not measured
original MIT
PreToolUse
23Hook
Runs before the agent uses a tool for Edit, Write, MultiEdit, NotebookEdit and Bash tool calls, executing repro-gate.sh. From cekura-ai/cekura-skills.
6 3d ago A
tokens not measured
original MIT
Hook
Runs when a tool call fails for mcp__cekura__.* tool calls, executing on-mcp-failure.sh. From cekura-ai/cekura-skills.
6 3d ago A
tokens not measured
original MIT