diff --git a/.claude-plugin/plugin.json b/.claude-plugin/plugin.json
index 1d657c0..aae4612 100644
--- a/.claude-plugin/plugin.json
+++ b/.claude-plugin/plugin.json
@@ -1,6 +1,6 @@
 {
   "name": "project-kit",
-  "version": "0.2.0",
+  "version": "0.2.1",
   "description": "Bootstrap a new project with Rody's standard setup (beads, uv, CLAUDE.md/AGENTS.md, CI) and research new tooling to evolve the kit",
   "keywords": [
     "scaffold",
diff --git a/catalog/setup-options.md b/catalog/setup-options.md
index e0e5b44..68e3fcb 100644
--- a/catalog/setup-options.md
+++ b/catalog/setup-options.md
@@ -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
+  `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
diff --git a/skills/research-setup/SKILL.md b/skills/research-setup/SKILL.md
index 2bb7ebd..f21daba 100644
--- a/skills/research-setup/SKILL.md
+++ b/skills/research-setup/SKILL.md
@@ -13,7 +13,7 @@ into concrete wiring in the kit. This keeps the setup evolving instead of frozen
 - **Profile / values** (from github-map): applied ML + data engineering with **own data and
   Peruvian regional domain** as the differentiator; reproducibility; agentic dev (Claude Code + beads).
 - **Already installed** (don't re-propose / weigh overlap): `superpowers`, `frontend-design`,
-  `github-map` plugins; `playwright` + `context7` MCP servers.
+  `github-map`, `claude-scholar`, `exa` plugins; `playwright` + `context7` MCP servers.
 
 ## Workflow