Playbooks · Skill authoring
Write and check an agent skill
Create a minimal discoverable skill in the open format and check its metadata and directory boundaries.
FREE AGENT SKILLS · MCP · AUTOMATION
Playbooks
Free public resources
Agent artifactsAgent authorizationAgent contextAgent evaluationAgent interoperabilityAgent memoryAgent observabilityAgent operationsAgent orchestrationAgent outputAgent recoveryAgent reliabilityAgent researchAgent runtimeAgent safetyAgent setupAgent stateAgent toolsAgent workflowsContent automationCrypto quant researchCrypto quant workflowsData automationDocument automationDocument and data automationMCP clientsMCP reliabilityMCP resourcesMCP serversMCP testingRetrievalSkill authoring
Playbooks · Agent setup
Enable and check a Gemini CLI skill
Check skill discovery locations, installation provenance, and activation in a session.
Playbooks · Agent tools
Define a validated function tool for an agent
Expose a narrowly scoped application function as a tool with an explicit parameter contract.
Playbooks · Agent safety
Add human approval to a side-effecting agent tool
Pause an agent run before an external change so a person can inspect and approve the exact action.
Playbooks · Agent safety
Validate inputs and final outputs in an agent workflow
Add testable rejection conditions for user input and final answers, distinguishing per-tool checks.
Playbooks · Agent orchestration
Have a manager agent call specialists and synthesize one answer
Delegate bounded subtasks to specialist agents while a manager retains ownership of the final answer.
Playbooks · Agent orchestration
Route a task to a specialist agent by explicit intent
Have a triage agent route matching requests to the right specialist and define who answers after transfer.
Playbooks · Agent safety
Expose only necessary context to a specialist during handoff
Trim or redact conversation history during handoff to reduce irrelevant and sensitive data exposure.
Playbooks · Agent recovery
Resume an interrupted agent task from saved run state
Save resumable run state so the same agent workflow can continue after interruption.
Playbooks · Agent state
Continue an agent conversation across requests with a persistent session
Associate a persistent session with the same user or task so later runs can use relevant history without mixing users.
Playbooks · Agent recovery
Distinguish agent final answers, refusals, and pending tool results
Handle completion, refusal, handoff, and pending tool calls by result type so intermediate events are not shown as answers.
Playbooks · Agent observability
Render an agent event stream incrementally and handle completion states
Map streamed agent events to stable UI states without misrendering internal tool events as final answers.
Playbooks · Agent evaluation
Test agent orchestration with a deterministic mock model
Verify application-owned tool dispatch, handoffs, and output handling without calling a live model service.
Playbooks · Agent observability
Trace model, tool, and handoff spans in an agent run
Correlate model calls, tool execution, guardrails, and handoffs in one workflow timeline for diagnosis.
Playbooks · MCP servers
Build a read-only MCP tool server for agent queries
Expose a narrowly scoped query tool so MCP clients can read application data but cannot change it.
Playbooks · MCP resources
Expose versioned read-only documents through MCP resource URIs
Expose documents as stable URI resources so clients can discover and read them without write access.
Playbooks · MCP servers
Define a stable error contract for MCP tools
Separate retryable failures, input rejections, and internal faults without exposing server secrets.
Playbooks · MCP clients
Configure OAuth authorization for a remote MCP client
Have an MCP client obtain a least-privilege token through the resource server’s declared flow and handle redirects safely.
Playbooks · MCP reliability
Make long-running MCP tool calls respond to cancellation
Propagate cancellation from the client to a long-running task, clean up resources, and stop unnecessary work.
Playbooks · MCP testing
Test MCP server discovery and tool calls locally
Connect an SDK test client to a local server and verify protocol discovery, input validation, and tool results.
Playbooks · MCP clients
Review MCP tools and side effects before agent use
Inspect names, descriptions, schemas, permissions, and side effects before exposing remote MCP tools to an agent.
Playbooks · Agent context
Control context and preserve key state in long agent tasks
Define compressible history and facts that must persist in long tasks to avoid context bloat or loss of constraints.
Playbooks · Agent memory
Retrieve relevant memories from completed agent sessions
Under user consent and retention rules, index completed sessions so later sessions can retrieve relevant context.
Playbooks · Agent state
Scope agent state by user, application, and invocation
Use explicit state namespaces to control which values persist across sessions and which are invocation-only.
Playbooks · Agent artifacts
Store large agent-generated files as versioned artifacts
Store reports, charts, or files as binary artifacts while keeping only the name and version in model context.
No matching entries.