imperal.io
  • How it works
  • Use cases
  • Technology
  • Pricing
Login
Webbee AIWebbee Codeimperal-mcpDevelopers & SDKSecurity & Vault
Add an app · Marketplace
imperal.ioICNLI Cloud OS

One OS for everything you run. Webbee is your agent inside it, she does the work, you stay in control.

Product
  • Webbee
  • Webbee Code
  • Imperal MCP
  • Marketplace
  • Use cases
  • Pricing
  • Open panel
Resources
  • Security
  • Technology
  • Success stories
  • Blog
Developers
  • Developers
  • Docs
  • Quick start
  • SDK reference
  • Developer portal
Company
  • About
  • Contact
  • Legal
Open protocol
  • ICNLI
  • SDK on GitHub
© 2026 Imperal, Inc. · ICNLI published at icnli.org under CC BY-SA 4.0
TermsPrivacy
  1. Home
  2. /
  3. Marketplace
  4. /
  5. Sweep
CO

Sweep Extension

📋 Compliance & Audit·by bluebeeweb·v0.1.0
Pay per use1-40 tok/action
Open in panel
0 capabilities0 actions0 installs

About this extension

Sweep Sustainability API connector: carbon footprints, emission factors, survey tracking, and ESG compliance health.

What you get

Every capability exposed by Sweep, metered per action and guarded by the two-step safety protocol.

  • connect_sweep_connectorwrite

    Typed function declaration. The platform reads ``chain_callable`` from the manifest to pick typed dispatch (direct ``app/func(args)`` call) vs. legacy ChatExtension LLM-router delegation. Defaults to ``True`` for ALL ``action_type`` values since v4.2.10, so the platform can deterministically execute the typed tool without giving the LLM a chance to summarise instead. ``effects`` declares the *intended* side-effect surface (``["create:note"]``, ``["delete:folder"]``, …). **Reserved / advisory:** the kernel does not currently consume it (no narrator / audit-ledger read today); it is retained as declared-intent metadata emitted into the manifest. ``id_projection`` declares the Pydantic params field that carries the resolved target id when this tool runs as a downstream chain step. Default empty - the platform falls back to a verb-prefix heuristic (``delete_note`` -> ``note_id``). Required for compound names where the heuristic produces a wrong field (``delete_notes_from_folder`` would naively yield ``notes_from_folder_id`` instead of ``folder_id``).

  • list_connectionsread

    Typed function declaration. The platform reads ``chain_callable`` from the manifest to pick typed dispatch (direct ``app/func(args)`` call) vs. legacy ChatExtension LLM-router delegation. Defaults to ``True`` for ALL ``action_type`` values since v4.2.10, so the platform can deterministically execute the typed tool without giving the LLM a chance to summarise instead. ``effects`` declares the *intended* side-effect surface (``["create:note"]``, ``["delete:folder"]``, …). **Reserved / advisory:** the kernel does not currently consume it (no narrator / audit-ledger read today); it is retained as declared-intent metadata emitted into the manifest. ``id_projection`` declares the Pydantic params field that carries the resolved target id when this tool runs as a downstream chain step. Default empty - the platform falls back to a verb-prefix heuristic (``delete_note`` -> ``note_id``). Required for compound names where the heuristic produces a wrong field (``delete_notes_from_folder`` would naively yield ``notes_from_folder_id`` instead of ``folder_id``).

  • disconnect_sweep_connectordestructive

    Typed function declaration. The platform reads ``chain_callable`` from the manifest to pick typed dispatch (direct ``app/func(args)`` call) vs. legacy ChatExtension LLM-router delegation. Defaults to ``True`` for ALL ``action_type`` values since v4.2.10, so the platform can deterministically execute the typed tool without giving the LLM a chance to summarise instead. ``effects`` declares the *intended* side-effect surface (``["create:note"]``, ``["delete:folder"]``, …). **Reserved / advisory:** the kernel does not currently consume it (no narrator / audit-ledger read today); it is retained as declared-intent metadata emitted into the manifest. ``id_projection`` declares the Pydantic params field that carries the resolved target id when this tool runs as a downstream chain step. Default empty - the platform falls back to a verb-prefix heuristic (``delete_note`` -> ``note_id``). Required for compound names where the heuristic produces a wrong field (``delete_notes_from_folder`` would naively yield ``notes_from_folder_id`` instead of ``folder_id``).

  • list_emissionsread

    Typed function declaration. The platform reads ``chain_callable`` from the manifest to pick typed dispatch (direct ``app/func(args)`` call) vs. legacy ChatExtension LLM-router delegation. Defaults to ``True`` for ALL ``action_type`` values since v4.2.10, so the platform can deterministically execute the typed tool without giving the LLM a chance to summarise instead. ``effects`` declares the *intended* side-effect surface (``["create:note"]``, ``["delete:folder"]``, …). **Reserved / advisory:** the kernel does not currently consume it (no narrator / audit-ledger read today); it is retained as declared-intent metadata emitted into the manifest. ``id_projection`` declares the Pydantic params field that carries the resolved target id when this tool runs as a downstream chain step. Default empty - the platform falls back to a verb-prefix heuristic (``delete_note`` -> ``note_id``). Required for compound names where the heuristic produces a wrong field (``delete_notes_from_folder`` would naively yield ``notes_from_folder_id`` instead of ``folder_id``).

  • get_emissionrecordread

    Typed function declaration. The platform reads ``chain_callable`` from the manifest to pick typed dispatch (direct ``app/func(args)`` call) vs. legacy ChatExtension LLM-router delegation. Defaults to ``True`` for ALL ``action_type`` values since v4.2.10, so the platform can deterministically execute the typed tool without giving the LLM a chance to summarise instead. ``effects`` declares the *intended* side-effect surface (``["create:note"]``, ``["delete:folder"]``, …). **Reserved / advisory:** the kernel does not currently consume it (no narrator / audit-ledger read today); it is retained as declared-intent metadata emitted into the manifest. ``id_projection`` declares the Pydantic params field that carries the resolved target id when this tool runs as a downstream chain step. Default empty - the platform falls back to a verb-prefix heuristic (``delete_note`` -> ``note_id``). Required for compound names where the heuristic produces a wrong field (``delete_notes_from_folder`` would naively yield ``notes_from_folder_id`` instead of ``folder_id``).

  • audit_emissionrecord_healthread

    Typed function declaration. The platform reads ``chain_callable`` from the manifest to pick typed dispatch (direct ``app/func(args)`` call) vs. legacy ChatExtension LLM-router delegation. Defaults to ``True`` for ALL ``action_type`` values since v4.2.10, so the platform can deterministically execute the typed tool without giving the LLM a chance to summarise instead. ``effects`` declares the *intended* side-effect surface (``["create:note"]``, ``["delete:folder"]``, …). **Reserved / advisory:** the kernel does not currently consume it (no narrator / audit-ledger read today); it is retained as declared-intent metadata emitted into the manifest. ``id_projection`` declares the Pydantic params field that carries the resolved target id when this tool runs as a downstream chain step. Default empty - the platform falls back to a verb-prefix heuristic (``delete_note`` -> ``note_id``). Required for compound names where the heuristic produces a wrong field (``delete_notes_from_folder`` would naively yield ``notes_from_folder_id`` instead of ``folder_id``).

Related extensions

More extensions in Compliance & Audit ready to enable in one click.

  • LogicGateConnect your own LogicGate (Risk Cloud) account (static API Token) to discover your configured Applications (Risk Register, Vendor Assessments, Incident Management, etc.), browse and manage Records...
  • MetricStreamConnect your own MetricStream tenant (static API Key) to browse the Risk register, review and manage Issues, check Controls testing status and Assessments, and get an aggregated risk-posture report.
  • VantaConnect your own Vanta organization (OAuth2 Client Credentials) to monitor compliance Tests/Controls/Frameworks, manage the Risk Register and Vendor risk inventory, review Policy documents and Pers...
  • WorkivaConnect your own Workiva workspace (bring your own OAuth2 API Grant Client ID/Secret from your Workiva administrator) to read and manage Sustainability Programs, Metrics and their Values, Dimension...

Ready to use Sweep?

Enable Sweep in your Imperal workspace in one click - 10,000 free starter credits included.

Open in panel