Claude Code v2.1.292 adds Agent effort control, plugin marketplace install flag, and prompt caching for mods
cliClaude Code 2.1.292 ships several agent-relevant additions: a new `effort` parameter on the Agent tool to control sub-agent effort level, `--marketplace <source>` on `claude plugin install` to add + install in one step, prompt caching in the `$.model.complete` mod API via `cache: true` blocks, a `prompt.autocomplete` hook, workflow agents exposed to `agent.spawn`, plus many security fixes (UNC path reads, sandbox escapes, tampered settings cache) and reliability fixes for bac
An agent can now dispatch sub-agents at a chosen effort level via the Agent tool's `effort` parameter, install plugins directly from a marketplace source in one command, and cache prompt/system blocks when mods call `$.model.complete`.
Notable: `effort` on Agent tool is a real orchestration primitive; prompt caching exposed to mods; MCP tool name >128 chars no longer breaks all requests; multiple security fixes around sandbox/UNC paths.
npm i -g @anthropic-ai/claude-code@2.1.292# Install a plugin and auto-add its marketplace in one shot
claude plugin install my-plugin --marketplace https://example.com/marketplace.json
# Spawn a sub-agent at a specific effort level from within Claude Code
# (inside a Claude Code session, the Agent tool now accepts `effort`)
# Agent(prompt="audit this repo", effort="high")
# Give overloaded (529) retries a longer base backoff
export CLAUDE_CODE_OVERLOADED_RETRY_BASE_DELAY_MS=5000
claude -p "summarize CHANGELOG.md"