Skill Claude CodeCodex
Marking Go agent methods as read-only for a side-effect-free guarantee and result caching. Use when the user wants a cacheable query method, a method that must not do remote side effects, or HTTP GET endpoints backed by a read-only method in a Go Golem project.
Skill Claude CodeCodex
Addressing multiple instances of a Go agent by identity. Use when the user wants many instances of an agent (one per user/tenant/session), a composite key/identity, or ephemeral/phantom instances in a Go Golem project.
Skill Claude CodeCodex
Fan out work to multiple parallel agents and collect results in a Go Golem project. Use when the user asks about parallel execution, fan-out/fan-in, spawning worker agents for parallel work, or aggregating results from multiple agents in a Go Golem project.
Skill Claude CodeCodex
Passes, derives, inspects, installs and revokes permission cards, and lends scope cards to agent calls, in a Go Golem project. Use when delegated authority must cross an RPC or tool boundary, or when an agent manages the cards in its wallet.
Skill Claude CodeCodex
Adding resource quotas to a Go Golem agent. Use when the user asks about rate limiting, resource quotas, quota tokens, golem.QuotaToken, reservations, throttling API calls, limiting concurrency, capacity limits, or splitting tokens between agents in a Go Golem project.
Skill Claude CodeCodex
Recurring / scheduled work in a Go agent. Use when the user wants a periodic task, a cron-like loop, a self-rescheduling agent, or to schedule a future invocation (of itself or another agent) in a Go Golem project.
Skill Claude CodeCodex
Configuring semantic retry policies for a Go Golem agent. Use when the user asks about retry policies, retry strategies, exponential backoff, transient error recovery, retry predicates, the retry subpackage, Named/Set/With, Policy or Predicate builders, jitter, MaxRetries, Clamp, AndThen, retrying the agent's own code…
Skill Claude CodeCodex
Scheduling a future invocation on a Go Golem agent from the CLI. Use when the user asks to schedule, delay, or plan a future agent method invocation in a Go Golem project.
Skill Claude CodeCodex
Scheduling a one-off future agent call from Go agent code, and canceling it. Use when the user wants a delayed invocation, a timed call for later, or to cancel a pending scheduled call in a Go Golem project.
Skill Claude CodeCodex
Creating ephemeral (stateless) agents in a Go Golem project. Use when the user wants a stateless agent, a fresh instance per invocation, no shared state between calls, or a request-handler style agent in a Go Golem project.
Skill Claude CodeCodex
Declares, implements, and calls streaming Go agent methods. Use for golem.AgentStream input or output, nested streams, ProduceStream, back-pressure, cancellation, forwarding, generated guest clients, or CLI streaming in a Go Golem project.
Skill Claude CodeCodex
Defines Golem tool middleware in Go. Use for validation, policy, auditing, caching, adapting one tool to another, or rewriting tool calls and results, either typed for one tool or universal for any tool, in a Go Golem project.
Skill Claude CodeCodex
Triggering a fire-and-forget invocation on a Go Golem agent from the CLI. Use when the user asks to trigger, enqueue, or fire-and-forget an agent method invocation in a Go Golem project.
Skill Claude CodeCodex
Unit tests a Go Golem component natively with go test, without building or deploying. Use when adding tests to a Go agent or tool, checking agent and tool definitions for mistakes, testing agent streams, or round-tripping values through Golem's schema.
Skill Claude CodeCodex
Waiting for external input using Golem promises in a Go Golem project. Use when the user asks about promises, waiting for external events, human-in-the-loop workflows, pausing an agent until an external signal, or suspending execution until data arrives from outside in a Go Golem project.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: