dcramer

Designs, reviews, and iterates on LLM agents and agent-like workflows. Use when asked to "design an agent", "review this agent", "improve our system prompt", "optimize prompts for caching", "improve tool calling", "reduce hallucinated tool calls", "add structured outputs", "decide if this should be multi-agent", "reduce false positives", "tune agent thresholds", or "build evals for this agent". Covers architecture choice, cache-friendly prompt templates, tool and schema design, runtime loops, trust boundaries, and eval-driven iteration.

Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.

Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete.

Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change.

Propose a new change with all artifacts generated in one step. Use when the user wants to quickly describe what they want to build and get a complete proposal with design, specs, and tasks ready for implementation.

Sync delta specs from a change to main specs. Use when the user wants to update main specs with changes from a delta spec, without archiving the change.

Use when manually QAing Peated changes across the API, CLI, or web UI; selecting smoke checks after code changes; using agent-browser for Peated flows; verifying local behavior beyond tests and lint.

Manage tasks via dex CLI. Use when breaking down complex work, tracking implementation items, or persisting context across sessions.

Create dex task from markdown planning documents (plans, specs, design docs, roadmaps)

Manage Gmail and Google Calendar with capability-backed auth and operations. Use when asked to check inbox, summarize emails, give a day at a glance, send an email, review calendar events, or schedule meetings.

Use when reviewing codebase health, finding structural issues, or auditing architecture quality before major changes

Use while implementing code changes, after a meaningful slice, to coordinate subagent review/fix/verify loops that preserve the core user or PR intent. Fix regressions, explicit requirement mismatches, validation gaps, and behavior-preserving cleanup; report unrelated improvements or out-of-intent behavior changes instead. Do not use for standalone reviews, brainstorming, or non-code iteration.

Use when transcript, commit, Sentry, GitHub, CI, review, or Garfield history suggests recurring failure patterns that might become hard deterministic checks. Analyze the evidence, reject judgment-based ideas, and output only mechanically enforceable lint, schema, CI, workflow, or deterministic test rules. Do not use for general code review, broad tool selection, or implementation unless explicitly asked.