release: v2026.03.29

Author: andre-inter-collab-llc

CHANGELOG.md

@@ -7,6 +7,38 @@ Same-day releases use a `.N` suffix (e.g., `2026.03.18.1`).
 
 A machine-readable mirror of this log is maintained in [versions.yaml](versions.yaml).
 
+## [2026.03.29] — 2026-03-29
+
+Script-first literature search workflow, Zotero attachment uploads, README Philosophy section, and infrastructure improvements.
+
+### Added
+
+- **Script-first search workflow**: `draft_*_search` and `run_search_script` tools added to all five database MCP servers (PubMed, OpenAlex, Semantic Scholar, Europe PMC, CrossRef). Formal literature searches now generate a standalone Python script for user review before execution, improving reproducibility and auditability.
+- **Shared search script engine**: `generate_search_script`, `execute_search_script`, and `generate_and_run_script` functions added to `rwa_result_store` shared library, powering the draft/run workflow across all servers.
+- **Zotero attachment upload**: `upload_attachment` tool added to zotero-server for uploading PDFs and other files to Zotero items via the Web API.
+- **PDF upload utility scripts**: `scripts/_upload_pdfs.py` and `scripts/_upload_pdfs2.py` for batch-uploading PDFs to Zotero library items.
+- **Search results backfill utility**: `scripts/backfill_search_exports.py` for regenerating Excel exports from existing `search_results.db` files.
+- **Bug report issue template**: `.github/ISSUE_TEMPLATE/bug-report.yml` for structured issue reporting.
+- **README Philosophy section**: Added project philosophy explaining RWA as a proof of concept, the case against platform lock-in, and the existing research tech stack, with a link to the LinkedIn launch post.
+- **Search strategy template update**: Added script-first workflow instructions to `templates/systematic-review/search-strategy.qmd`.
+- **Agent instructions for script-first workflow**: Updated `systematic-reviewer` and `research-orchestrator` agents with the mandatory draft-then-approve search protocol, including allowed direct MCP tool usage and DB-as-source-of-truth rules.
+- **Global copilot instructions**: Added Literature Search Protocol section enforcing the script-first, DB-as-source-of-truth workflow across all agents.
+
+### Changed
+
+- **MCP server project path handling**: Improved error messages and validation for `project_path` parameters across PubMed, OpenAlex, Semantic Scholar, Europe PMC, and CrossRef servers.
+- **API documentation**: Updated `docs/api-setup-guide.md`, `docs/database-access.md`, and `docs/getting-started.md` with clarified instructions.
+- **Search runners**: Improved `search_runners.py` in the shared library for better script execution handling.
+
+### Fixed
+
+- **Chat parser**: Handle non-string content in thinking blocks to prevent parsing errors in chat export.
+
+### Removed
+
+- Obsolete `search_results.db` files from sample project directories (data now lives in project-specific paths).
+- Stale `prisma-flow.json` from sample project (replaced by PRISMA tracker server workflow).
+
 ## [2026.03.18] — 2026-03-18
 
 Initial public release of the Research Workflow Assistant.

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "research-workflow-assistant"
-version = "2026.03.18"
+version = "2026.03.29"
 description = "AI-powered research workflow assistant with MCP servers for academic database access, ICMJE-compliant"
 readme = "README.md"
 license = {file = "LICENSE"}

requirements-lock.txt

@@ -1,5 +1,5 @@
 # Research Workflow Assistant — Environment Snapshot
-# Generated: 2026-03-18T20:48:37.441941+00:00
+# Generated: 2026-03-29T15:01:34.090120+00:00
 # Python: 3.13.12
 # Platform: win32
 #
@@ -8,58 +8,25 @@
 #   .venv/Scripts/activate  (or source .venv/bin/activate)
 #   pip install -r requirements-lock.txt
 
-annotated-types==0.7.0
-anyio==4.12.1
-attrs==25.4.0
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=bibliography_manager&subdirectory=mcp-servers\bibliography-manager
-certifi==2026.2.25
 cffi==2.0.0
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=chat_exporter&subdirectory=mcp-servers\chat-exporter
-click==8.3.1
-colorama==0.4.6
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=crossref_server&subdirectory=mcp-servers\crossref-server
-cryptography==46.0.5
-et_xmlfile==2.0.0
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=europe_pmc_server&subdirectory=mcp-servers\europe-pmc-server
-h11==0.16.0
-httpcore==1.0.9
-httpx==0.28.1
-httpx-sse==0.4.3
-idna==3.11
-iniconfig==2.3.0
-jsonschema==4.26.0
-jsonschema-specifications==2025.9.1
-mcp==1.26.0
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=openalex_server&subdirectory=mcp-servers\openalex-server
-openpyxl==3.1.5
-packaging==26.0
-pluggy==1.6.0
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=prisma_tracker&subdirectory=mcp-servers\prisma-tracker
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=project_tracker&subdirectory=mcp-servers\project-tracker
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=pubmed_mcp_server&subdirectory=mcp-servers\pubmed-server
+charset-normalizer==3.4.4
+contourpy==1.3.3
+cryptography==46.0.4
+cycler==0.12.1
+fonttools==4.59.2
+git-filter-repo==2.47.0
+kiwisolver==1.4.9
+matplotlib==3.10.6
+numpy==2.3.3
+packaging==25.0
+pandas==2.3.2
+pdfminer.six==20251230
+pdfplumber==0.11.9
+pillow==11.3.0
 pycparser==3.0
-pydantic==2.12.5
-pydantic-settings==2.13.1
-pydantic_core==2.41.5
-Pygments==2.19.2
-PyJWT==2.11.0
-PyMuPDF==1.27.2
-pytest==9.0.2
-pytest-asyncio==1.3.0
-python-dotenv==1.2.2
-python-multipart==0.0.22
-pywin32==311
-PyYAML==6.0.3
-referencing==0.37.0
-rpds-py==0.30.0
-ruff==0.15.5
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=rwa_result_store&subdirectory=mcp-servers\_shared
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=rwa_shared&subdirectory=mcp-servers\_shared
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=semantic_scholar_server&subdirectory=mcp-servers\semantic-scholar-server
-sse-starlette==3.3.2
-starlette==0.52.1
-typing-inspection==0.4.2
-typing_extensions==4.15.0
-uvicorn==0.41.0
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=zotero_local_server&subdirectory=mcp-servers\zotero-local-server
--e git+https://github.com/andre-inter-collab-llc/research-workflow-assistant.git@2c81163f8c02896fa63935a4060f103872ba47e8#egg=zotero_server&subdirectory=mcp-servers\zotero-server
+pyparsing==3.2.4
+pypdfium2==5.3.0
+python-dateutil==2.9.0.post0
+pytz==2025.2
+six==1.17.0
+tzdata==2025.2

versions.yaml

@@ -25,3 +25,28 @@ releases:
       fixed: []
       removed: []
       security: []
+
+  - version: "2026.03.29"
+    date: "2026-03-29"
+    summary: "Script-first literature search workflow, Zotero attachment uploads, README Philosophy section, and infrastructure improvements"
+    changes:
+      added:
+        - "Script-first search workflow: draft_*_search and run_search_script tools added to all five database MCP servers (PubMed, OpenAlex, Semantic Scholar, Europe PMC, CrossRef)"
+        - "Shared search script engine: generate_search_script, execute_search_script, and generate_and_run_script in rwa_result_store"
+        - "Zotero attachment upload: upload_attachment tool in zotero-server"
+        - "PDF upload utility scripts: scripts/_upload_pdfs.py and scripts/_upload_pdfs2.py"
+        - "Search results backfill utility: scripts/backfill_search_exports.py"
+        - "Bug report issue template: .github/ISSUE_TEMPLATE/bug-report.yml"
+        - "README Philosophy section with LinkedIn launch post link"
+        - "Search strategy template update for script-first workflow"
+        - "Agent instructions for script-first workflow in systematic-reviewer, research-orchestrator, and global copilot-instructions"
+      changed:
+        - "Improved error messages and project path validation across all database MCP servers"
+        - "Updated API documentation in docs/api-setup-guide.md, docs/database-access.md, docs/getting-started.md"
+        - "Improved search_runners.py in shared library"
+      fixed:
+        - "Handle non-string content in thinking blocks for chat parser"
+      removed:
+        - "Obsolete search_results.db files from sample project directories"
+        - "Stale prisma-flow.json from sample project"
+      security: []