From 6a90e43067b090a4b5d676dd10dd666a12e39908 Mon Sep 17 00:00:00 2001 From: Amr Gharbeia Date: Mon, 27 Apr 2026 13:22:55 -0400 Subject: [PATCH] docs: sync gtd.org roadmap with new v0.3.0-v1.0.0 version scheme - Renamed v0.3.0 to include HITL - Renamed v0.4.0 to include Git Workflows - Shifted Creator + Architect to v0.6.0 - Inserted new v0.5.0 (Interactive Actuation) - Inserted new v0.7.0 (Visual Grounding) and v0.8.0 (Evaluation Harness) before v1.0.0 SOTA Parity. --- gtd.org | 53 +++++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 49 insertions(+), 4 deletions(-) diff --git a/gtd.org b/gtd.org index 9859b4f..fffaff3 100644 --- a/gtd.org +++ b/gtd.org @@ -146,6 +146,13 @@ Roadmap basis: Evolutionary roadmap from README.org. Working backwards from SOTA - Rule 11: Org as thinking medium - document investigations in prose - Rule 12: Engineering decision audit trail - document root cause, tradeoffs - Added to opencortex-contrib/skills/org-skill-engineering-standards.org +***** DONE Perform comprehensive architectural review and evolution strategy [2026-04-27 Mon] + - Identified hidden gaps: Org-mode round-trip, sandboxing vulnerabilities, and GC scaling. + - Defined "Structural AST Editing" and "Reflection Loops" as core strategic requirements. + - Captured findings in [[file:notes/opencortex-architectural-evolution.org][opencortex-architectural-evolution.org]]. +***** DONE Fix API drift in opencortex-contrib [2026-04-27 Mon] + - Standardized legacy keywords (:neuro/:symbolic) to new harness standard (:probabilistic/:deterministic). + - Updated 16 skills in opencortex-contrib for kernel compatibility. **** DONE 4. Core Skills Consolidation [2026-04-23 Thu] - Merged lisp-validator + lisp-repair → org-skill-lisp-utils.org @@ -177,7 +184,7 @@ Roadmap basis: Evolutionary roadmap from README.org. Working backwards from SOTA **** TODO 3. Direct Lisp-to-Terminal Actuation ***** TODO Refactor the =:cli= actuator to use native TUI rendering -*** PHASE: EVENT ORCHESTRATION (v0.3.0) +*** PHASE: EVENT ORCHESTRATION + HITL (v0.3.0) :PROPERTIES: :ID: proj-orchestration-v0-3-0 :END: @@ -215,7 +222,7 @@ Unified control plane: hooks + cron + routing in one skill. Deep project underst ***** TODO /- prefix for command mode ***** TODO Commands defined in Org-mode -*** PHASE: LONG-HORIZON PLANNING (v0.4.0) +*** PHASE: LONG-HORIZON PLANNING + GIT WORKFLOWS (v0.4.0) :PROPERTIES: :ID: proj-planning-v0-4-0 :END: @@ -246,7 +253,21 @@ Multi-step task mastery, structured tracking with failure handling and course co - Navigate, clock time, refile, archive - Uses org-element + org-id -*** PHASE: CREATOR + ARCHITECT (v0.5.0) +*** PHASE: INTERACTIVE ACTUATION & ENVIRONMENT STEWARDSHIP (v0.5.0) +:PROPERTIES: +:ID: proj-actuation-v0-5-0 +:END: +Interactive terminal sessions and autonomous dependency management. + +**** TODO 0. Interactive PTY Actuator +***** TODO Stream long-running process output to the context window (e.g., `npm run dev`, REPLs) +***** TODO Implement async interrupt control (Ctrl+C emulation) + +**** TODO 1. The Environment Steward +***** TODO Autonomously detect missing dependencies (e.g., "Command not found") +***** TODO Propose an installation command and retry the failed action + +*** PHASE: CREATOR + ARCHITECT + GTD (v0.6.0) :PROPERTIES: :ID: proj-creator-v0-5-0 :END: @@ -286,13 +307,37 @@ Agent bootstraps itself: creates skills autonomously, designs projects from PRDs - Detect orphaned nodes - PARA/Zettelkasten maintenance +*** PHASE: VISUAL GROUNDING & MCP BRIDGE (v0.7.0) +:PROPERTIES: +:ID: proj-vision-v0-7-0 +:END: +Multimodal visual interaction and ecosystem-wide tool compatibility. + +**** TODO 0. Computer Use / Vision +***** TODO Allow the agent to request host OS or browser screenshots +***** TODO Analyze UI and issue precise X/Y coordinate click/type commands via an X11/Wayland bridge + +**** TODO 1. MCP Gateway Bridge +***** TODO Build a Lisp-native client for the Model Context Protocol +***** TODO Connect OpenCortex to external tools and data sources + +*** PHASE: THE EVALUATION HARNESS (v0.8.0) +:PROPERTIES: +:ID: proj-eval-v0-8-0 +:END: +Automated benchmarking to mathematically prove the agent's reasoning capabilities. + +**** TODO 0. SWE-Bench Harness +***** TODO Automated pipeline that clones repositories and feeds GitHub issues +***** TODO Track multi-step resolution trajectory, run tests, and score success + *** PHASE: SOTA PARITY (v1.0.0) :PROPERTIES: :ID: proj-sota-v1-0-0 :END: Feature-complete agent competitive with commercial agents. All borrowed concepts reimplemented in pure Lisp. -All features from v0.2.0 through v0.5.0 combined, verified, and tested end-to-end. +All features from v0.2.0 through v0.8.0 combined, verified, and tested end-to-end. | Area | Parity Target | |------|--------------|