Tutorial / template 2026
Beyond Keywords: June15 archival research method slides (Madore)
The selected June15 Beyond Keywords presentation documents user-scoped archival questions, model-selected read-only tools and attributed historical synthesis; the original slides-collection identity is retained.
L3Multi-step tool-using agentic workflow (workflow level, not a quality rating) Source: Repository-documented Outcome: Not empirically evaluated here
Research applicability
A concrete source-described humanities research use of AI-mediated collection access and historical research methodology.
Getting started
Untested editorial guidance: read the pinned primary documentation, choose one public/synthetic research question and inspect original source records/limitations before any separate tool trial. Keep final scholarly interpretation with the researcher. No trial was executed here.
Maintenance
Current source commit/metadata saved2026-10-19; long-term support or availability not established.
License
Not independently audited; repository availability does not establish rights to every embedded archival item or provider data.
Workflow & human oversight
Scoped slide-described L3 for the selected June15 Beyond Keywords archival research method: user question/depth, model-selected read-only tools and scoping/searching/reading/triangulation/attributed synthesis. It is not a level for the entire presentation collection, a runtime audit, enforced expert signoff or L4 scientific experimentation.
Workflow level is not an evidence score; higher is not better evidence.
Approach
The selectedBeyondKeywords deck describes brief/extended modes and scoping, systematic searching with null-result logging, deep reading, triangulation and attributed synthesis. It presents read-only MCP access and an agent skill as methodology; no implementation/runtime audit performed.
Limitations
Documentation only; no repository code, model calls, installation, data processing or experiment executed. Primary API/raw source retrieval does not validate implementation, security, performance or scientific outcomes. One selected deck plus manifest and selectedREADME context; no audit of allseven decks or embedded190line skill. Saved-run screenshots/chats were not inspected and illustrative search slices are not validated retrieval experiments. The deck’s blanket no-AI-at-query-time statement is qualified by its explicit live Gemini query-embedding tool. No inherited IWAC outcomes, collection counts, universal multilingual accuracy or enforced expert signoff claimed. The workshop/collective-position-paper material elsewhere in the collection is outside this selected June15 method scope.
Reproducibility
Commit-pinned primary documentation saved; no tool execution, result reproduction or safety/effectiveness verification.
Adoption requirements
Read the public slides; any later underlying tool trial needs its own current setup/provider/data-rights review. The presentation is not the installed MCP server or full embeddedskill.
Comments
No comments yet.
Add a comment
Comments are held for moderation before appearing. Critical or corrective comments are welcome — especially evidence that an entry overstates what has been verified.