agentleFS
Sign inStart free
QuixiAIQuixiAI/Hexis/skills/installed/integration-connector-setup/SKILL.md

integration-connector-setup

Skill623 starsChanged 10 months agoMITView on GitHub

Start, configure, verify, inspect, and revoke first-class connectors including Slack, Telegram, Signal, Notion, Spotify, Home Assistant, Weather, and Trello

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. About 0 are common, 3 are distinctive, and 11 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 11

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this when the user asks to connect Slack, Telegram, Signal, Notion, Spotify, Home Assistant, Weather, or Trello, or to inspect …

    line 13
  2. Distinctive:

    Treat setup as an in-conversation flow with exact next steps from the DB connector manifest.

    line 18
  3. Distinctive:

    Never ask the user to paste bot tokens, app tokens, passwords, or API secrets into chat.

    line 19
  4. Very specific:

    Use connect_gmail from the Gmail connector skill for Gmail OAuth.

    line 20
  5. Very specific:

    Use connect_life_integration for Notion, Home Assistant, Weather, and Trello.

    line 21
  6. Very specific:

    Use connect_spotify for Spotify PKCE.

    line 22
  7. Very specific:

    After config is written or the user says env vars are set, call verify_channel_integration so the DB records the connection only when the …

    line 25
  8. Very specific:

    Call integration_setup_status for the requested connector or for all connectors if the user asks what can be connected.

    line 29
  9. Very specific:

    Call start_integration_setup with the least capabilities matching the user's request.

    line 30
  10. Very specific:

    If the user provides channel settings, call configure_channel_integration only with env var names for token fields and non-secret …

    line 31
  11. Very specific:

    Call verify_channel_integration after the env/config values should be available.

    line 32
  12. Very specific:

    If the user asks to import history, call start_connector_backfill with the smallest useful scope — Slack needs a channel_id.

    line 34
  13. Very specific:

    Use connector_backfill_status for progress, and control_connector_backfill only when the user asks to pause, resume, or cancel a job.

    line 35
  14. Very specific:

    Use complete_spotify_connection only when the automatic local callback did not finish.

    line 37

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Integration Connector Setup

Use this when the user asks to connect Slack, Telegram, Signal, Notion, Spotify, Home Assistant, Weather, or Trello, or to inspect available connectors.

Principles

  • Treat setup as an in-conversation flow with exact next steps from the DB connector manifest.
  • Never ask the user to paste bot tokens, app tokens, passwords, or API secrets into chat. Token fields must be env var names.
  • Use connect_gmail from the Gmail connector skill for Gmail OAuth; this skill covers manual/pairing channel connectors.
  • Use connect_life_integration for Notion, Home Assistant, Weather, and Trello. Pass only environment variable names for secret fields. Never search ambient environment variables or paste secret values into tool arguments.
  • Use connect_spotify for Spotify PKCE. Its app client ID is non-secret, but an environment value may be used only when the user explicitly names that environment variable. Register the exact loopback redirect URI returned by Hexis; Spotify does not accept localhost.
  • Be explicit when a capability is planned rather than available for a given provider.
  • Historical backfill is local memory ingestion and needs explicit user approval, even when the provider access is read-only. Queue it only for a connected connector, after the user asks to import history.
  • After config is written or the user says env vars are set, call verify_channel_integration so the DB records the connection only when the channel worker's config truth resolves.

Flow

  1. Call integration_setup_status for the requested connector or for all connectors if the user asks what can be connected.
  2. Call start_integration_setup with the least capabilities matching the user's request.
  3. If the user provides channel settings, call configure_channel_integration only with env var names for token fields and non-secret allowlists or URLs.
  4. Call verify_channel_integration after the env/config values should be available.
  5. Tell the user to start or restart hexis-channels only if verification succeeded and the adapter is not already running.
  6. If the user asks to import history, call start_connector_backfill with the smallest useful scope — Slack needs a channel_id; Telegram, Signal, and Twitter/X take a local export/archive path (export_path/import_path).
  7. Use connector_backfill_status for progress, and control_connector_backfill only when the user asks to pause, resume, or cancel a job.
  8. For an everyday-life API connector, call its connect tool once without missing fields to surface the provider-specific setup card; verify only after the user chooses the credential references or default location.
  9. Use complete_spotify_connection only when the automatic local callback did not finish. Use revoke_life_integration when the user asks to disconnect one of these providers.

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.

Check your setup, free