← StateCourse AICONTENT HISTORY

Update to StateCourse AI

Snapshot Oct 31, 2026 · 01:14 UTC · version 1.0.0

Collection source: downloaded plugin package.

WHAT CHANGED

First saved snapshot

No earlier snapshot is available to establish a change.

Save StateCourse AI and check future changes in Following.

Up to 10 plugins free · no card required.

Compare saved observations

Download comparison JSON
Full technical diff · 0 changed fields
Full snapshot data
{
  "description": "Assemble an internal NMLS or state education submission package from authorized project documents, including a manifest, evidence index, readiness sheet, missing-item register, filename controls, and checksums. Use when a provider team asks to prepare, organize, review, or export a submission package without filing it.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 539
    }
  ],
  "name": "submission-package-builder",
  "skill_md_contents": "---\nname: submission-package-builder\ndescription: Assemble an internal NMLS or state education submission package from authorized project documents, including a manifest, evidence index, readiness sheet, missing-item register, filename controls, and checksums. Use when a provider team asks to prepare, organize, review, or export a submission package without filing it.\n---\n\n# Submission Package Builder\n\n1. Confirm the agency, filing type, applicant, deadline, controlling checklist, and authorized project.\n2. Require Provider entitlement.\n3. Pin the controlling requirement set and retrieve the project inventory.\n4. Use `build_submission_package` to assemble only authorized artifacts. Preserve source identity, version, status, and confidentiality classification.\n5. Produce:\n   - package status/read-me;\n   - item-level readiness table;\n   - final upload manifest;\n   - evidence index;\n   - missing and blocked items;\n   - checksum register;\n   - authorized-human action list.\n6. Run export DLP. Exclude secrets, credentials, answer keys, internal reviewer materials, unrelated customer data, and superseded drafts unless explicitly retained as history.\n7. Label the package `UNSIGNED — UNSUBMITTED` unless verified external evidence states otherwise.\n\nNever submit, email, upload, sign, attest, pay, or represent the package as accepted. Never turn a missing official form into an AI-created substitute. If the user asks for external action, stop at a precise handoff.\n\n"
}

SHA-256 of public snapshot: ea601d5445efac8f6d83b1507be8827b6e52ea70d8b7214a3143f57583788cc1