agentleFS
AiqbeeAiqbee/aiqbee-ai/skills/brain/SKILL.md

brain

Skill7 starsChanged 8 months agoMITView on GitHub

List, inspect, create, and manage your Aiqbee brains — configure settings, manage user access, and switch brain context.

25 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use research.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • NVIDIASkill

    research

    Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.

    Also in 4 projects that use this

The file

Show its rules and text

We split this skill into the rules it gives an agent

19 rules. 0 are common, 5 are distinctive, and 14 are very specific to this project.

  • 0

    Common

    Most of the 20 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 14

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If arguments don't match any command, treat them as a brain name to search for and show its info.

    line 28
  2. Very specific:

    Call aiqbee_list_brains().

    line 34
  3. Very specific:

    Call aiqbee_list_brain_templates().

    line 38
  4. Very specific:

    Call aiqbee_get_brain_info(brain_id).

    line 42
  5. Very specific:

    After creation, show the new brain's ID and mcpUrl, and ask if the user wants to switch to it (run the set action).

    line 51
  6. Very specific:

    Call aiqbee_update_brain(brain_id, ...).

    line 55
  7. Very specific:

    Call aiqbee_delete_brain(brain_id).

    line 59
  8. Very specific:

    If the argument is a name (not a UUID), call aiqbee_list_brains() to resolve the brain ID

    line 65
  9. Very specific:

    Read the project's .mcp.json file

    line 66
  10. Very specific:

    Update the aiqbee entry URL from https://mcp.aiqbee.com/mcp to https://mcp.aiqbee.com/brain/{brain_id}/mcp

    line 67
  11. Very specific:

    Write the updated .mcp.json

    line 68
  12. Very specific:

    Update the aiqbee entry URL to https://mcp.aiqbee.com/mcp

    line 76
  13. Very specific:

    Call aiqbee_list_users(brain_id).

    line 82
  14. Very specific:

    Call aiqbee_grant_access(email, brain_id, access_level).

    line 86
  15. Very specific:

    Call aiqbee_revoke_access(email, brain_id).

    line 90
  16. Distinctive:

    If no exact match, try a partial (substring) match

    line 97
  17. Distinctive:

    If multiple brains match, always show the list and ask the user to pick one — never assume

    line 98
  18. Distinctive:

    If no match, tell the user and show available brains

    line 99
  19. Distinctive:

    Write operations also require the brain's AllowMCPEditing setting to be enabled.

    line 119

“Common” is measured: kept by at least 6 in 10 of the 20 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Aiqbee Brain Management

You are a brain management assistant for the Aiqbee MCP server. Parse the user's arguments and execute the appropriate action.

Command routing

Parse $ARGUMENTS and route to the correct action:

CommandAction
(empty) or listList all brains
templatesList brain templates
info [name-or-id]Get brain details
create <name> [--template <id>] [--org] [--no-edit]Create a brain
update <name-or-id> [--name <n>] [--desc <d>]Update a brain
delete <name-or-id>Delete a brain (ask for confirmation first)
set <name-or-id>Switch MCP config to a brain-specific URL
unsetSwitch MCP config back to the brain-agnostic URL
access [name-or-id]List users with access
grant <email> [Read|ReadWrite|Owner] [name-or-id]Grant user access
revoke <email> [name-or-id]Revoke user access

If arguments don't match any command, treat them as a brain name to search for and show its info.

Action details

list (default)

Call aiqbee_list_brains(). Display results as a formatted table with columns: Name, Access Level, MCP Editing, ID. Highlight the brain that matches the current MCP URL if using a brain-specific endpoint.

templates

Call aiqbee_list_brain_templates(). Display as a table with: Name, Description, ID. Mention that users can pass a template ID to create.

info

Call aiqbee_get_brain_info(brain_id). If the argument looks like a name rather than a UUID, first call aiqbee_list_brains() to find the matching brain ID. Show all brain metadata.

create

Call aiqbee_create_brain(name, ...).

  • --template <id>: Pass as brain_template_id
  • --org: Set is_personal=false
  • --no-edit: Set allow_mcp_editing=false

After creation, show the new brain's ID and mcpUrl, and ask if the user wants to switch to it (run the set action).

update

Call aiqbee_update_brain(brain_id, ...). If the argument is a name, resolve to ID first via aiqbee_list_brains(). Only pass fields the user explicitly provided.

delete

Call aiqbee_delete_brain(brain_id). Always ask the user for confirmation before deleting. Warn that this permanently deletes all neurons, types, relationships, and access.

set

Switch the MCP config to a brain-specific URL. Steps:

  1. If the argument is a name (not a UUID), call aiqbee_list_brains() to resolve the brain ID
  2. Read the project's .mcp.json file
  3. Update the aiqbee entry URL from https://mcp.aiqbee.com/mcp to https://mcp.aiqbee.com/brain/{brain_id}/mcp
  4. Write the updated .mcp.json
  5. Tell the user to restart their MCP client for the change to take effect

unset

Switch the MCP config back to the brain-agnostic URL. Steps:

  1. Read the project's .mcp.json file
  2. Update the aiqbee entry URL to https://mcp.aiqbee.com/mcp
  3. Write the updated .mcp.json
  4. Tell the user to restart their MCP client for the change to take effect

access

Call aiqbee_list_users(brain_id). If no brain specified, use the current brain from the MCP URL. Display as a table with: Name, Email, Access Level, Creator.

grant

Call aiqbee_grant_access(email, brain_id, access_level). Default access level is "Read" if not specified. Confirm the grant was successful.

revoke

Call aiqbee_revoke_access(email, brain_id). Confirm the revocation was successful.

Brain ID resolution

Many commands accept a brain name or ID. To resolve a name to an ID:

  1. Call aiqbee_list_brains()
  2. Try an exact case-insensitive match first
  3. If no exact match, try a partial (substring) match
  4. If multiple brains match, always show the list and ask the user to pick one — never assume
  5. If no match, tell the user and show available brains

Destructive actions (delete, update, set, grant, revoke): require an exact match or explicit user selection from the match list. Never auto-resolve a partial match for these commands.

Endpoint context

The Aiqbee MCP server has two endpoint patterns:

  • Brain-agnostic: https://mcp.aiqbee.com/mcp — requires brain_id on each tool call
  • Brain-specific: https://mcp.aiqbee.com/brain/{id}/mcp — all tools target that brain automatically

The set / unset commands modify .mcp.json to switch between these.

Access levels

LevelCapabilities
ReadView brain, neurons, relationships
ReadWriteRead + create/update/delete neurons and relationships
OwnerReadWrite + manage brain settings, neuron types, and user access

Write operations also require the brain's AllowMCPEditing setting to be enabled.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.