SeaShell keeps AI edits safe by gating every tool call against allow, ask, and deny rules before it runs. Each file edit computes a unified diff you approve before anything is written, and changes are snapshotted as git patches so they can be rolled back. In 2026 that means automated work stays auditable and under your control.
Permission modes
Before any tool runs, SeaShell evaluates it against your rules and resolves to allow, ask, or deny. Unmatched actions default to ask, so nothing surprising runs without a prompt.
This permission model applies to every tool call, from editing a file to running a shell command. You can broaden rules for trusted actions and tighten them for risky ones, keeping a clear boundary around what the agent may do automatically versus what needs your confirmation.
Plan before edits
SeaShell drafts a plan in a read-only plan agent and asks before switching to a build agent to implement. You see the shape of a change before it touches your tree.
Separating planning from editing means you can refine the approach while no files are at risk. Once you approve, SeaShell implements the plan, and each edit still passes through the diff-and-permission step. Planning and permissions reinforce each other.
Diffs and rollback
Each file edit computes a unified diff that is surfaced for approval before writing. SeaShell also snapshots changes as git patches, so a change can be diffed and rolled back.
Reviewing a real diff before it lands is the core safety primitive: you approve exactly what changes. If something is wrong after the fact, the snapshot lets you restore. Together, permissions, plans, diffs, and snapshots make automated edits auditable end to end.
Frequently asked questions
- How does SeaShell keep AI edits safe?
- Every tool call is permission-gated against allow, ask, and deny rules, and every edit is shown as a unified diff before it is written.
- Can I control what the agent runs automatically?
- Yes. You set rules that resolve each action to allow, ask, or deny; unmatched actions default to ask.
- Does SeaShell plan before editing?
- Yes. SeaShell drafts a plan in a read-only plan agent and asks before switching to a build agent to implement.
- Can I undo an SeaShell change?
- Yes. SeaShell snapshots file changes as git patches, so edits can be diffed and rolled back.
Try SeaShell in your terminal
The terminal-native AI coding agent. Free core, single binary.
Install SeaShellRelated guides
Atlas with Qwen3 32B (Ollama) in 2026
Explore Atlas with Qwen3 32B (Ollama), a powerful 32B dense model for complex coding tasks. Self-hosted and free, it offers 40K tokens for deep reasoning.
Atlas for Next.js in 2026
Adopt Atlas, the terminal-native AI coding agent, for Next.js development in 2026. Enhance productivity across App Router, server components, and API routes with secure, reviewable AI assistance.
Atlas with Magistral Medium in 2026: Deep Reasoning for Complex Code
In 2026, Atlas developers can leverage Magistral Medium for deep, multi-hop root-cause analysis across services, utilizing its 128,000 token context window.
Atlas with GLM-5.2 in 2026: A Developer's Guide
Explore Atlas with GLM-5.2, Z.ai's 2026 flagship model. Leverage its 1M token context and open-weights flexibility for agentic coding at $1.40/Mtok input. Understand its tradeoffs.
Atlas with Mistral NeMo 12B (Ollama) in 2026
Explore Atlas with Mistral NeMo 12B (Ollama), a free, self-hosted 12B model offering a 128K practical context window for local AI coding in 2026.
Atlas with North Mini Code in 2026
Explore Atlas with North Mini Code, Cohere's 2026 code-specialist model. Benefit from its 256,000 token context, 64,000 token output, and $0 pricing for large refactors.
Atlas with Claude Opus 4.8 in 2026
Discover Atlas with Claude Opus 4.8, Anthropic's top coding model released in May 2026. Leverage its 1M token context window and advanced reasoning for complex multi-file refactors at $5/$25 per Mtok.
Atlas with GLM-5.1 in 2026
Explore Atlas with GLM-5.1, a frontier model from Z.ai offering strong reasoning and a 200,000 token context. Understand its $1.40/$4.40 pricing and tradeoffs for developers in 2026.