Projects
Source curator
Import and organize sources with correct provenance, Project placement, and reusable taxonomy.
What this skill teaches
Every source uses the Prism ingestion path that matches its real provenance.
Fixed source kinds remain distinct from reusable user tags.
Canonical URLs, Project placement, classification, and source metadata remain inspectable.
Workflow
- 1
Call get_library_taxonomy when content categories, source kinds, tags, or supported upload formats affect the decision.
- 2
List or search the destination Project first to avoid duplicate sources and tags.
- 3
Use import_library_web_source for a readable canonical URL. It verifies extractability before creating a Library item, so report a rejected blocked, login-gated, binary, or low-content URL rather than retrying it. Use the YouTube transcript tools for a supported video source, upload_library_file_source for a local file, and create_library_note only for original or user-supplied text without a canonical URL.
- 4
Apply sourceKind and reusable tags deliberately, then re-read the created source metadata.
- 5
Use update_library_file_metadata for supported metadata edits rather than recreating a source.
Tool routing
Inspect the current classification and destination before creating anything.
Preserve a readable page's canonical URL and website provenance. The import preflight rejects pages Prism cannot extract without creating a failed source.
Preview availability, then save a transcript-backed video source.
Create original notes or upload supported local bytes without pretending they came from the web.
Maintain reusable labels and editable metadata after the source exists.
Safety boundaries
- Never substitute a note or upload when a readable canonical URL was requested and the web import fails.
- File upload does not fetch arbitrary remote files. Provide local bytes through the supported base64 or data URL input.
- Do not treat sourceKind as a free-form tag or rewrite imported source bodies that Prism marks as non-editable.
SKILL.md source
Public Markdown you can inspect, copy, download, or adapt.
---
name: prism-source-curator
description: "Use when an AI client needs to import, classify, tag, or edit private Prism Project sources while preserving provenance."
---
# Source curator
Import and organize sources with correct provenance, Project placement, and reusable taxonomy.
## Before you start
Copy this guidance into ChatGPT, Claude, Codex, Grok, or another client that accepts custom instructions. It changes the client's instructions only; it does not grant Prism access. Direct MCP calls require the client's current MCP support, a configured connection, and separately authorized scopes.
Scopes this workflow may need:
- `knowledge:read`
- `library:write`
- `library:edit`
The installed skill does not grant these permissions. Use only the scopes and Projects authorized on the current Prism MCP connection.
## Workflow
1. Call get_library_taxonomy when content categories, source kinds, tags, or supported upload formats affect the decision.
2. List or search the destination Project first to avoid duplicate sources and tags.
3. Use import_library_web_source for a readable canonical URL. It verifies extractability before creating a Library item, so report a rejected blocked, login-gated, binary, or low-content URL rather than retrying it. Use the YouTube transcript tools for a supported video source, upload_library_file_source for a local file, and create_library_note only for original or user-supplied text without a canonical URL.
4. Apply sourceKind and reusable tags deliberately, then re-read the created source metadata.
5. Use update_library_file_metadata for supported metadata edits rather than recreating a source.
## Tool routing
- `get_library_taxonomy`, `list_library_sources`: Inspect the current classification and destination before creating anything.
- `import_library_web_source`: Preserve a readable page's canonical URL and website provenance. The import preflight rejects pages Prism cannot extract without creating a failed source.
- `preview_youtube_transcript_source`, `import_youtube_transcript_source`: Preview availability, then save a transcript-backed video source.
- `create_library_note`, `upload_library_file_source`: Create original notes or upload supported local bytes without pretending they came from the web.
- `create_library_tag`, `tag_library_sources`, `update_library_file_metadata`: Maintain reusable labels and editable metadata after the source exists.
## Expected result
- Every source uses the Prism ingestion path that matches its real provenance.
- Fixed source kinds remain distinct from reusable user tags.
- Canonical URLs, Project placement, classification, and source metadata remain inspectable.
## Boundaries
- A skill does not grant Prism access. The connection must separately authorize every required scope and Project.
- Call only tools verified callable by the current AI host. Use discover_prism_actions for granular actions. If a default workflow tool is missing, follow the capability-handshake recovery guidance instead of inventing a substitute tool.
- Read before writing. Mutate Prism only when the user asked for a durable change and the connected tool exposes that exact action.
- Keep retrieved evidence separate from analysis. Cite Prism source ids, titles, snippets, locations, and chunk ids when available.
- Never substitute a note or upload when a readable canonical URL was requested and the web import fails.
- File upload does not fetch arbitrary remote files. Provide local bytes through the supported base64 or data URL input.
- Do not treat sourceKind as a free-form tag or rewrite imported source bodies that Prism marks as non-editable.