Guides
Token limits and workflow playbooks.
Practical guides for Claude Code, Cursor, and Codex, tied to AgenKit features and docs.
Token limits
Reduce Claude Code token usage
Practical ways to spend fewer tokens in Claude Code sessions, including local memory, scoped reads, and workflow discipline.
Token limits
Claude Code usage limit reached: what to do
When Claude Code hits a usage or rate limit, pause, reduce context, and resume with a tighter workflow instead of repeating the same expensive prompt.
Token limits
Cursor credits running out
Stretch Cursor credits by narrowing agent scope, using local memory, and running structured workflows instead of open-ended refactors.
Token limits
Codex usage limits
Work within Codex CLI limits by invoking focused skills, keeping tasks small, and pairing Codex with local memory from AgenKit.
Workflow
Claude Code subagents explained
How specialist subagents differ from one general assistant, and how AgenKit maps eleven roles to Claude Code.
Workflow
Test-driven development with Claude Code
Use the Test Engineer agent and /agenkit build phase to keep RED-GREEN-REFACTOR discipline in Claude Code projects.
Workflow
Claude Code code review workflow
Run a two-stage code review (spec compliance, then quality) with the Code Reviewer and Debugger agents in AgenKit.
Workflow
Claude Code memory: CLAUDE.md vs codebase index
Compare hand-written CLAUDE.md notes with AgenKit Local Memory: deterministic indexes, query commands, and committed digests.