[DOCS] Agent view docs omit `claude agents --dangerously-skip-permissions` dispatch support added in v2.1.142

Resolved 💬 1 comment Opened May 14, 2026 by coygeek Closed Jun 1, 2026

Documentation Type

Incorrect/outdated documentation

Documentation Location

https://code.claude.com/docs/en/agent-view

Section/Topic

"Permission mode, model, and effort" and "Settings, plugins, and MCP servers" in the agent view guide

Current Documentation

The docs currently say:

To set defaults for every session you dispatch from agent view, pass any of --permission-mode, --model, or --effort when opening it: ``bash claude agents --permission-mode plan --model opus --effort high ` Passing --permission-mode, --model, or --effort to claude agents` requires Claude Code v2.1.142 or later. Earlier versions reject these flags with an unknown-option error.

Later on the same page, the flag table says:

| --settings <file-or-json> | Override settings for agent view and dispatched sessions | | --add-dir <path> | Grant file access to an additional directory | | --plugin-dir <path> | Load a plugin from a local directory | | --mcp-config <file-or-json> | Load MCP servers from a config file or JSON string | | --strict-mcp-config | Use only the MCP servers from --mcp-config, ignoring other MCP configuration |

But the CLI reference separately documents:

| --dangerously-skip-permissions | Skip permission prompts. Equivalent to --permission-mode bypassPermissions. See permission modes for what this does and does not skip |

What's Wrong or Missing?

The v2.1.142 changelog says claude agents gained support for several flags when configuring dispatched background sessions, including --dangerously-skip-permissions, but the agent view page only documents three of the permission/model/effort defaults directly and a separate table for settings, plugins, MCP, and extra directories.

That leaves a gap for the bypass-permissions form of the new agent-view dispatch configuration:

A. The agent view page does not explicitly document claude agents --dangerously-skip-permissions

The page explains --permission-mode, --model, and --effort, but not the direct bypass-permissions flag even though the CLI reference still exposes it as a first-class flag.

B. Users reading the agent view page can infer the wrong limitation

Because the page says "pass any of --permission-mode, --model, or --effort" and then lists a separate pass-through table that also omits --dangerously-skip-permissions, it reads as though the only permission-related flag supported by claude agents is --permission-mode.

Suggested Improvement

Update the agent view page to document the direct bypass flag alongside the other v2.1.142 dispatch defaults.

Minimum fix:

  • Expand the "Permission mode, model, and effort" section to mention that claude agents --dangerously-skip-permissions is also supported as the bypass-permissions equivalent of --permission-mode bypassPermissions.
  • Add a short example such as claude agents --dangerously-skip-permissions --model opus.
  • If the recommended form is --permission-mode bypassPermissions instead, say that explicitly and note whether --dangerously-skip-permissions is still accepted for parity with the top-level CLI.

Impact

Medium - Makes feature difficult to understand

Additional Context

Affected Pages:

| Page | Line(s) | Context |
|------|---------|---------|
| https://code.claude.com/docs/en/agent-view | 304-332 | Documents the new claude agents dispatch defaults and pass-through flags, but omits --dangerously-skip-permissions |
| https://code.claude.com/docs/en/cli-reference | 63 | Confirms --dangerously-skip-permissions remains a documented first-class CLI flag |
| https://code.claude.com/docs/en/changelog | v2.1.142 entry | Release notes say claude agents added --add-dir, --settings, --mcp-config, --plugin-dir, --permission-mode, --model, --effort, and --dangerously-skip-permissions for configuring dispatched background sessions |

Total scope: 3 pages affected

This appears to be a v2.1.142 docs sync gap rather than a product bug: the feature is referenced in the changelog and the underlying CLI flag is documented, but the main agent view guide does not spell out the claude agents form.

View original on GitHub ↗

This issue has 1 comment on GitHub. Read the full discussion on GitHub ↗