{% extends "base.html" %} {% block title %}LASSO — Edit Profile: {{ profile.name }}{% endblock %} {% block breadcrumb %}
  • {% endblock %} {% block content %} {% if error %} {% endif %}
    {{ csrf_token() }}

    Basic Information

    General settings that identify this profile and define its inheritance.

    {% if source == 'builtin' %}

    Built-in profiles cannot be renamed. Use "Duplicate & Edit" to create a copy.

    {% endif %}

    Inherit settings from a base profile. Your changes override the base.

    Help your team understand when to use this profile.

    Allowed Commands

    Control which shell commands the AI agent can execute. An allowed list is safer -- only the commands you check will be permitted.

    Check the commands you want to allow (or block, depending on mode).

    Core -- file viewing and search

    {% for cmd in ['ls', 'cat', 'head', 'tail', 'grep', 'find', 'wc', 'sort', 'diff', 'echo', 'test'] %} {% endfor %}

    Development -- build and version control

    {% for cmd in ['python3', 'pip', 'git', 'node', 'npm', 'make', 'cargo', 'go'] %} {% endfor %}

    Network -- download and transfer use with caution

    {% for cmd in ['curl', 'wget'] %} {% endfor %}

    File operations

    {% for cmd in ['mkdir', 'cp', 'mv', 'touch', 'tar', 'zip', 'unzip'] %} {% endfor %}

    Add commands not listed above, separated by commas.

    Kill commands that run longer than this.

    Network Access

    Control whether the AI agent can access the internet and which sites it can reach. Offline is the safest option.

    The agent can only connect to these domains. Click a domain to remove it.

    {% for domain in profile_data.network.allowed_domains %} {{ domain }} × {% endfor %}

    Common database ports are blocked by default to prevent accidental data access.

    {% set db_ports = [ (3306, 'MySQL'), (5432, 'PostgreSQL'), (27017, 'MongoDB'), (6379, 'Redis'), (1433, 'SQL Server'), (1521, 'Oracle'), (9042, 'Cassandra'), (5984, 'CouchDB'), ] %} {% for port, label in db_ports %} {% endfor %}

    Resource Limits

    Prevent runaway processes by limiting memory, CPU, and process count. Lower values are safer but may limit the agent's ability to build large projects.

    512MB 16GB
    10% 100%

    Maximum number of simultaneous processes the agent can spawn.

    Audit Settings

    Every action inside the sandbox can be recorded so you can review what the AI did. These settings control how detailed that recording is.

    Cancel
    {% endblock %} {% block scripts %} {% endblock %}