codebase-analyzerAnalyzes codebase implementation details. Call the codebase-analyzer agent when you need to find detailed information about specific components. As always, the more detailed your request prompt, the better! :)
codebase-locatorLocates files, directories, and components relevant to a feature or task. Call `codebase-locator` with human language prompt describing what you're looking for. Basically a "Super Grep/Glob/LS tool" — Use it if you find yourself desiring to use one of these tools more than once.
codebase-pattern-findercodebase-pattern-finder is a useful subagent_type for finding similar implementations, usage examples, or existing patterns that can be modeled after. It will give you concrete code examples based on what you're looking for! It's sorta like codebase-locator, but it will not only tell you the locatio
implementerInternal to tce's step commands (`/tce:work`, `/tce:implement`, `/tce:quickfix`) — not for direct use. The implementation step of the tce workflow: implements exactly one phase of an approved tce plan — changes, the phase's Automated checks with a bounded fix loop, the plan's implementation log and
plan-challengerInternal to tce's step commands (`/tce:work`, `/tce:plan`, `/tce:quickfix`) — not for direct use. Checks a freshly written plan against a closed checklist (criteria verifiable, tests per phase, no contradiction with stack or research, intent criteria covered, phases independently verifiable) and, in
plan-compliance-checkerVerifies an implementation against a fixed list of acceptance/success criteria, from a fresh context. Call it at the end of implementation to get one verdict per criterion (met / not met / cannot verify from diff / needs human verification) against the diff — nothing else. Pass it only the criteria
plannerInternal to tce's step commands (`/tce:work`, `/tce:plan`, `/tce:quickfix`) — not for direct use. The planning step of the tce workflow: plans a ticket or a ticketless intent from the intent and its research, questions first — returns the open questions before writing anything, then writes and commi
presenterInternal to tce's step commands (`/tce:work`, `/tce:plan`, `/tce:implement`, `/tce:quickfix`) and `/tce:intent` — not for direct use. Writes the human-facing summary of one finished tce artifact — a plan brief or its push-back delta, an intent, a dossier — from a fresh read-only context, shaped by t
researcherInternal to tce's step commands (`/tce:work`, `/tce:research`, `/tce:quickfix`) — not for direct use. The research step of the tce workflow: researches a ticket or a ticketless intent against the live codebase (fanning out to the locator, analyzer, pattern-finder, thoughts and web agents) and writes
thoughts-analyzerThe research equivalent of codebase-analyzer. Use this subagent_type when wanting to deep dive on a research topic. Not commonly needed otherwise.
thoughts-locatorDiscovers relevant documents in thoughts/ directory (We use this for all sorts of metadata storage!). This is really only relevant/needed when you're in a reseaching mood and need to figure out if we have random thoughts written down that are relevant to your current research task. Based on the name
verifierInternal to tce's step commands (`/tce:work`, `/tce:implement`, `/tce:quickfix`) — not for direct use. The verification step of the tce workflow: verifies a fully implemented tce plan — runs every test suite, runs the plan-compliance gate, then attempts every item the gate could not settle with the
web-search-researcherDo you find yourself desiring information that you don't quite feel well-trained (confident) on? Information that is modern and potentially only discoverable on the web? Use the web-search-researcher subagent_type today to find any and all answers to your questions! It will research deeply to figure