> ## Documentation Index
> Fetch the complete documentation index at: https://docs.traycer.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Artifact workflow

> How agents use Traycer's built-in skills to plan, build, and review work through artifacts.

<Note>
  This is the Desktop workflow. The IDE extension is a separate product; see
  the [IDE Extension docs](/extension).
</Note>

A **Task** is the container for a piece of work. It holds agents, files, the git diff, and [artifacts](/panels/artifacts): specs, tickets, stories, and reviews. Traycer ships a set of skills that tell agents how to plan, build, and review through those artifacts.

## How Artifacts Appear

Artifacts are written when something should outlive the conversation or be put in front of you. You can create them from the [Artifacts](/panels/artifacts) panel, and an agent can write one during a turn.

An agent writes an artifact only when the work calls for one. If none is written, the coding work can still finish and the context stays in the chat transcript. To be sure you get an artifact, ask for it, or run the skill that produces it.

## Built-In Skills

For a large feature, a typical order is brief, core flows, tech plan, tickets, execute, review, and walkthrough. Skip any step the work doesn't need.

### Plan

| Skill | What it does |
| - | - |
| `traycer-epic-brief` | Frames a new initiative: what it is, who it affects, why it matters now, and what's in scope. Writes a short brief. |
| `traycer-core-flows` | Maps the user journeys, states, and UX decisions at product level, after clarifying them with you. |
| `traycer-tech-plan` | Works through the architecture and load-bearing technical decisions with you, then writes plan artifacts. |
| `traycer-ticket-breakdown` | Turns the plan into ticket artifacts with their dependencies, and iterates until you approve the set. |
| `traycer-revise-requirements` | Carries a significant requirement change through the product, technical, and ticket artifacts, in that order. |
| `traycer-artifact-critique` | Pressure-tests a plan or other artifact for gaps, contradictions, and risks. Writes a review artifact. |

### Build

| Skill | What it does |
| - | - |
| `traycer-execute` | Orchestrates implementation of the tickets. It orders them into batches by dependency, starts one child agent per ticket, and reviews each batch against the plan. Tickets that pass are marked done; small problems get a fix-up ticket; product drift stops the run and comes back to you. |
| `traycer-implement` | Guides an agent that implements a ticket handed to it by another agent, such as `traycer-execute`: when to ask, when to decide, and what to report back. You rarely run it yourself. |
| `traycer-autobuild` | Runs a long, mostly hands-off build. The agent drafts a high-level spec, a grading rubric, and a way to verify the work, and asks you to approve them. Then a builder agent and a separate evaluator agent iterate on each sprint until it passes. |

### Review

| Skill | What it does |
| - | - |
| `traycer-review` | Reviews a code change for correctness and quality. Writes a review artifact with prioritized findings. |
| `traycer-changeset-walkthrough` | Writes a guide to reviewing completed work: what changed, why, the review order, and what still needs verifying. The agent offers one after substantial work and writes it once you agree. |

### Decide and Maintain

| Skill | What it does |
| - | - |
| `traycer-debate` | Runs a structured debate between agents with different perspectives. You approve the perspectives and models, and it ends with a written synthesis. |
| `traycer-housekeeping` | Audits the Traycer-created worktrees on the host, recommends which to remove, and deletes only the ones you approve. |

## Where Skill Output Lands

| Skill | Output |
| - | - |
| `traycer-review`, `traycer-artifact-critique` | A **review** artifact. |
| `traycer-ticket-breakdown` | One **ticket** per piece of work, under a `tickets` folder. |
| `traycer-changeset-walkthrough` | A **spec** under a `changeset-walkthroughs` folder. |
| `traycer-debate` | Artifacts under `debates/<topic>/`: one per round and perspective, and a final synthesis. |
| `traycer-autobuild` | Artifacts under `autobuild/<build>/`: the spec, the rubric, and one per sprint. |

Folders are relative to the Task's artifacts. See [Artifacts on Disk](/panels/artifacts#artifacts-on-disk).

## Run a Skill

In the chat composer, type `$` or `/` to open the picker; both open the same list. Pick a skill, then write your prompt after it. You can attach several skills to one prompt.

```text theme={null}
$traycer-epic-brief We want the mobile app to work offline. Frame the initiative.
```

```text theme={null}
$traycer-tech-plan Plan the sync engine for offline mode, based on the core flows.
```

```text theme={null}
$traycer-execute Implement the offline-mode tickets.
```

```text theme={null}
$traycer-review Review the changes on this branch.
```

Some coding agents also start a skill on their own when a request matches its description. Traycer doesn't run the steps as an automatic pipeline and doesn't fail a turn when a skill is skipped. Picking the skill yourself is the reliable way to run it.

To add skills of your own, see [Skills](/settings/providers/skills).

## Feedback Through Comments

Comment on the artifacts as you review them, then ask an agent to address the open threads. The agent reads each thread, updates the artifact, and resolves the threads it has handled. See [Comments and Agents](/panels/comments#comments-and-agents).

## Related

* [Tasks and workspace folders](/concepts/tasks-and-workspace-folders)
* [Artifacts](/panels/artifacts)
* [Agents](/panels/agents)
* [Agent-to-agent communication](/concepts/agent-to-agent)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.