# Single-Agent Workflow
Even in single-agent environments, Decapod provides a rigorous structure that prevents common agentic errors and ensures high-quality delivery (see [Agent-First Architecture](../concepts/agent-first.md)).
## The Standard Loop
1. **Orientation:** The agent reads `AGENTS.md` and initializes its session (see [CLI Reference](../reference/cli.md#core-operations)).
- `decapod session acquire`
2. **Intent Capture:** The agent identifies its task and formalizes the intent (see [Explicit Intent](../concepts/intent.md)).
- `decapod todo claim --id <id>`
- `update specs/INTENT.md` (see [Artifacts Reference](../reference/artifacts.md))
3. **Workspace Entry:** The agent moves into an isolated environment (see [Workspace Sandboxing](../concepts/workspaces.md)).
- `decapod workspace ensure`
4. **Implementation:** The agent performs the work within the workspace.
5. **Validation:** The agent verifies the change against project policy.
- `decapod validate`
6. **Completion:** The agent generates proof and marks the task as done (see [Proof & Validation](../concepts/proof.md)).
- `decapod todo done --id <id> --validated`
## Key Benefits
- **Safe Iteration:** The agent works on a dedicated branch, meaning it can't accidentally break the main build while experimenting.
- **Forced Specification:** The agent is forced to think about "Why" and "How" before writing code, leading to more coherent designs.
- **Verifiable Outcome:** The human operator receives a PR with a `decapod validate` pass, providing confidence in the change.