Repository navigation
docs: record literature-review tooling research option #9
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -64,6 +64,42 @@ it is ours to publish. | |
| A deliberate exception to wiring only adopted tools: the prompt asks the question and lists the | ||
| options, but installs no HF tooling. Ships in plugin 0.2.0. | ||
|
|
||
| ## 2026-09-11 — tooling for a verified literature review (+ BibTeX for a LaTeX report) | ||
|
|
||
| Need: find, verify and cite ~9 recent papers (≤5 years) for a course deliverable, and produce a | ||
| `.bib` for a LaTeX report. Constraints: every DOI/venue/figure must be checkable (the deliverable | ||
| is graded on source quality), no paid API, and it must work from the CLI. | ||
|
|
||
| - **Candidate**: `claude-scholar` plugin — already installed (`openalex`, `doi-bibtex`, | ||
| `check-refs`, `arxiv-metadata` skills) | ||
| - What it does: OpenAlex queries, DOI → BibTeX, and a reference checker for LaTeX `.bib` files. | ||
| - Fit: strong. No key, free, and `check-refs` closes the loop against the written report — which | ||
| matches the reproducibility value better than a search-only tool. | ||
| - Maintenance: installed plugin, in use. | ||
| - Verdict: **adopt** — primary tool. Used for the PC1 review in `cursos/concurrente` (R0SEWT/concurrente#5). | ||
| - **Candidate**: direct REST calls to OpenAlex + Crossref + `doi.org` content negotiation (curl) | ||
| - What it does: same data, one layer down; `api.crossref.org/works?query.bibliographic=` turned | ||
| out to be **better than OpenAlex for CS venues** (found the ICPP/IEEE Access papers that | ||
| OpenAlex relevance search buried), and `Accept: application/x-bibtex` on `doi.org` gives BibTeX. | ||
| - Verdict: **adopt** as the fallback whenever a skill's results look thin. Worth knowing that | ||
| OpenAlex `title_and_abstract.search` silently misses records with no stored abstract. | ||
| - **Candidate**: an academic MCP server — ScholarMCP (https://github.com/lstudlo/scholarmcp), | ||
| `openags/paper-search-mcp` (https://github.com/openags/paper-search-mcp), | ||
| `oksure/openalex-research-mcp` (https://github.com/oksure/openalex-research-mcp) | ||
| - What they do: wrap OpenAlex/Crossref/Semantic Scholar/arXiv search, PDF ingestion, citation export. | ||
| - Fit: heavy overlap with `claude-scholar` + `exa`, both already installed. PDF ingestion is the | ||
| only real gap, and it did not come up: the deliverable needs metadata and DOIs, not full texts. | ||
| - Verdict: **reject** for the template — revisit only if a project needs bulk PDF ingestion. | ||
| - **Candidate**: `exa` plugin (installed) for discovery | ||
| - Verdict: **trial**, complementary — good at finding the *non-indexed* material (vendor blogs, | ||
| a paper accepted but not yet in a DOI registry), weak for bibliographic metadata. | ||
| It runs on a metered Exa API key, so under the no-paid-API constraint it stays optional — | ||
| never part of the required verification path. | ||
| - **Friction found**: the **dblp API is behind a bot check** (`dblp.org/search/publ/api` returns an | ||
| Anubis HTML challenge, HTTP 200). Don't script it; use Crossref for CS-venue coverage instead. | ||
| - **Wiring**: none. All adopted pieces are already installed or plain `curl`; nothing to add to | ||
|
Owner
Author
There was a problem hiding this comment. The entry records The next research-setup run doesn't weigh overlap with claude-scholar or exa, and may re-propose the academic MCP servers this entry just rejected for overlapping with them. "Wiring: none" should include that one-line update to the skill. Automated review (Claude Code, AI-generated). Validate before acting.
Owner
Author
There was a problem hiding this comment. Fixed in 688863d: added to the research-setup list and bumped the plugin to 0.2.1. (Claude Opus 5.5, on behalf of Rody) |
||
| `templates/`. Recorded here so the next literature task starts with Crossref in hand. | ||
|
|
||
| ## 2026-06-02 — an MCP for SQLite | ||
|
|
||
| Need: an MCP server to let Claude query a local SQLite DB. Constraint: must fit the local data | ||
|
|
||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
PR #5 inserts its entry at the same spot, directly under
<!-- entries below -->.git merge-treeof the two branches reports a conflict.Whichever merges second hits a conflict. The log is documented as "Newest first", so the 2026-09-11 entry has to sit below 2026-09-24. A quick resolution that keeps "theirs on top" (for example, merging #9 after #5) puts the entries out of date order.
Automated review (Claude Code, AI-generated). Validate before acting.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Fixed in 688863d: #9 is now stacked on #5, with this entry below 2026-09-24. Merge #5 first. (Claude Opus 5.5, on behalf of Rody)