jmylchreest

assess-findings

Triage static analysis findings, assess merit, and accept noise or irrelevant items

autopilot

Full autonomous execution - keeps working until all tasks are verified complete

code-search

Search code symbols, find function calls, and analyze codebase

context-usage

Analyze current session context and token usage from OpenCode SQLite database

debug

Systematic debugging workflow for tracking down bugs and issues

decide

Formal decision-making interview for architectural choices

design

Technical design and architecture for implementation

forget

Soft-delete memories by adding/removing the forget tag, or hard-delete outdated memories

memorise

Capture key learnings, patterns, gotchas, or context as persistent memories

patterns

Analyze codebase patterns, detect anti-patterns, and surface static analysis findings

plan-swarm

Socratic planning interview for swarm decomposition

recall

Search memories and decisions to answer questions about past learnings, choices, and context

survey

Explore codebase structure, entry points, tech stack, hotspots, and call graphs

swarm-status

Inspect a running swarm — show the agent tree, current tools, halts/pauses, and recent task/message activity for the orchestrator's own swarm.

verify

Active QA verification - full validation before completion

worktree-resolve

Resolve and merge git worktrees after swarm completion

build-fix

Fix build, lint, and type errors

docs

Update documentation to match implementation

git

Git operations and worktree management

implement

TDD implementation - make failing tests pass

perf

Performance analysis and optimization workflow

review

Code review and security audit

swarm

Parallel agents with SDLC pipeline per story

semgrep

Run Semgrep security and code quality analysis