ORE Studio Sprint 18 – Release Notes

• Build and publish user manual PDF via CMake: A deploy_manual CMake target compiles user_manual.org to a PDF via Emacs/LaTeX.
• Resolve RFL complexity issues: Permanently resolve MSVC C1202 and Clang expression-nesting failures in ores.qt.api via TU splits, two-phase parsing, and increased compiler limits.
• SQLite CLI quality-of-life setup: Committed .sqliterc gives headed, box-drawn, non-wrapping output with NULL markers and query timing from the sqlite3 shell.
• Lock down uppercase UUID invariant: All :ID: properties and [[id:UUID]] links in doc/ must be uppercase; all code paths that mint UUIDs produce uppercase by construction.
• Remove unused GitHub Actions workflows: Delete the four dormant Gemini workflows; PR review stays via the gemini-code-assist app.

• Add support for running ORE samples: End-to-end execution of ORE sample input files from within ORE Studio: workspace promoted to a full entity, data import session backend and UI, and ORE results surfaced in the UI.

• ores.compass repository compass tool: New ores.compass Python project: Locate (current version, sprint, in-flight stories and tasks), Search (semantic doc retrieval from org-roam data), and Scaffold (add stories, tasks, sprints, recipes via codegen).
• ores.compass Locate — temporal orientation: compass where reports current version, sprint, and in-flight stories and tasks computed from the documentation graph.
• ores.compass Scaffold — agile authoring: compass add story/task/sprint/recipe creates agile artefacts from the CLI.
• Track PRs per task and surface in compass where: PRs table in task docs; compass where --prs fetches live PR status via gh.
• Add org-roam index links to compass.org folder entries: Every folder entry in doc/compass.org links to its index file.
• compass cross-worktree status (global where): One command shows every git worktree's branch, story, task, and open PR, preventing parallel-agent collisions.
• compass goto: start a unit of work in one command: compass goto fetches main, creates the branch, and scaffolds the linked story and task in a single step.
• compass goto –kind flag: Add --kind hotfix to compass goto for hotfix/<slug> branches; update runbook, recipe, and docs.
• Per-sprint capture workflow: individual files and compass integration: Per-sprint captures/ folder with compass capture command replacing the single overloaded capture.org.
• Fix product-backlog bucket names in scripts and captures: Update regenerate_backlog_indexes.py, compass.py, and captures to use next/deferred instead of the old near/far names.

• Product backlog refinement: backlog-refiner and sprint-planner skills with companion recipes for structured backlog housekeeping and sprint planning.
• Sprint health review — System 2 analysis: sprint-reviewer skill writes a structured health-check section into sprint.org: goal alignment, load, PR velocity, story/task balance, focus signal, and verdict.
• Sprint health charts: Daily charts of PRs, commits, line churn, and story completions rendered as PNGs into the sprint page.
• Refactor story task lists to a status table: * Tasks section in story docs converted from bullet lists to State/Start/End/Description tables.
• Sprint theme grouping: * Stories in each sprint.org grouped under theme subheadings.
• Hotfix handling: Hotfix runbook, branching recipe update, and Hotfixes section in the sprint template.

• Runbook support: New runbook document type as a named composition of recipes for repeatable multi-step operations.
• Improve LLM runbook and skill tooling: Add compass skill type, update skill-creation recipe, create run-runbook skill.
• Document deduplication recipe: Generalised recipe for merging duplicate docs; applied to the duplicate sprint health review runbooks.
• PlantUML integration: recipe and codegen scaffold: PlantUML how-to recipe, compass add diagram scaffold, and standard licence header convention.
• Document ORE Studio as an LLM-first system: LLM-first philosophy page and Claude Code tool page; update llm.org index.
• Port Tufte visualisation skill into ORE Studio: tufte-viz SKILL and two Tufte reference docs converted to org-mode and catalogued.
• Port domain concept notes to knowledge documents: Domain notes converted to knowledge documents under doc/knowledge/domain/.

• System model readability: Layer blurbs and table-based component inventory added to the system model; architecture diagram linked.
• Document NATS: technology overview, system integration, and recipes: NATS knowledge page, system model section, per-component subject tables, and NATS recipes.
• Tooling documentation: Tooling section added to knowledge.org; developer scripts doc; overhaul of compass, sql, lisp, and codegen overviews.
• Add Emacs and org-roam knowledge pages: Emacs, org-roam, and Zettelkasten knowledge pages; ores.lisp expanded; orientation updated.
• Improve cybernetics documentation: VSM overview page, expanded Cybernetic Levels with context and limitations, compass link added.
• Inline org-roam export into ores.lisp: Vendor org-roam-export.el into ores.lisp so the CI site build is self-contained.

• Doc format cleanup: migrate v1 docs and remove v2 branding: Eliminate all "v2" branding from the document toolchain; audit and migrate remaining v1 docs.
• ORE Studio Emacs dashboard: Env-aware emacs-dashboard console replaces the broken per-mode ores.lisp approach where scripts from one checkout bled into another.
• Move .build-.el scripts to ores.lisp*: Relocate five build scripts from the project root into projects/ores.lisp/src/ with the ores- prefix convention.

Dual-axis bar chart. PRs (left axis) and commits (right axis) per day. A high commits-to-PR ratio may indicate scope creep.

Lines added (green) and deleted (red) per day. Building work produces mostly additions; refactoring produces a mix.

Hours from PR open to merge, one bar per PR. Long bars indicate review bottlenecks.

Line chart tracking stories marked DONE during the sprint. Steady upward slope is healthy; plateauing signals a stall.