ticlawk 0.1.16-dev.21 → 0.1.16-dev.22

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ticlawk",
-  "version": "0.1.16-dev.21",
+  "version": "0.1.16-dev.22",
   "description": "Local connector that links agent harnesses (Claude Code, Codex, OpenClaw, opencode, Pi) to the Ticlawk mobile app.",
   "type": "module",
   "main": "ticlawk.mjs",

package/src/runtimes/_shared/goal-task-protocol.mjs

@@ -119,19 +119,13 @@ Goal authority overlay: responsible for this conversation's goal and tasks
 - You are responsible for driving the conversation toward its goal, not only replying to isolated messages.
 - Maintain or infer the current goal from the direct ask, charter, dashboard/briefing quote, task board, and conversation context.
 ${buildGoalSetupRules()}
-- Run the goal loop:
-  1. Evaluate current facts against the goal.
-  2. Identify the concrete gap, if any.
-  3. Decompose the gap into concrete tasks when useful.
-  4. Execute the next task yourself when appropriate, or create/assign a task when coordination is needed.
-  5. Check whether the result closes the gap.
-  6. Return to evaluating current facts against the goal.
-- At goal-loop boundaries, explicitly write a private goal loop check: current facts, goal/success criterion, gap, next action, and stop/continue decision. Do this when entering the goal loop and after checking whether a task result closes the gap. This is for your execution discipline, not a chat message. Do not send the private goal loop check through ${BRAND_NAME} unless the owner explicitly asks for that analysis.
-- Stop the loop only when there is no meaningful gap, progress is blocked because the owner needs to provide input/resources/permission/confirmation/decision, or progress depends on an external/time-based wait.
-- If there is no gap, say so briefly. If the goal is complete, report completion.
-- If user input, resources, permission, confirmation, or a decision is needed, publish a briefing with one clear bundled request to the owner.
-- If progress depends on an external or time-based future state and no agent or owner can act now, use a reminder or explicit resume condition rather than silently stalling. Do not use reminders to defer an executable next step or an owner decision/request.
-- When reviewing returned task work, evaluate the evidence against the task and current goal before marking it done or using it in reports. If the task is complete, update task state, then immediately run the private goal loop check against the updated facts before dashboard, MEMORY.md, briefing, or wrap-up updates.
+- Run the goal loop as a state machine. At each goal-loop boundary, explicitly write a private goal loop check in normal assistant output with: current facts, goal/success criterion, state, gap, next action, and stop/continue decision. This is required execution trace, not a chat message. Never send this internal state through ${BRAND_NAME}; \`ticlawk message send\` should contain only the effective external message: a task instruction, result/evidence, blocker, owner request, or final status.
+- State \`clear_gap\`: there is a concrete gap the agent system can act on now. Execute the next task yourself when appropriate, or create/assign a group task when coordination is needed. If you delegate substantive work, use the task board and send a clean task instruction.
+- State \`unclear_gap\`: the gap or success criterion is not concrete enough to act on. Gather the smallest missing context yourself, create an investigation task, or ask one bundled clarification in the group/briefing depending on who can answer it.
+- State \`owner_blocked\`: progress needs owner input, resources, permission, confirmation, or a decision. Publish one briefing with a clear bundled request, then stop until the owner responds.
+- State \`external_wait\`: progress depends on an external or time-based future state and no agent or owner can act now. Use a reminder or explicit resume condition, then stop. Do not use reminders to defer an executable next step or owner decision/request.
+- State \`returned_task\`: returned work is ready for review. Evaluate the evidence against the task and current goal before accepting it. If complete, update task state, then immediately run another private goal loop check against the updated facts before dashboard, MEMORY.md, briefing, or wrap-up updates. If incomplete, keep or return the task with a clean redo/blocker instruction.
+- State \`complete\`: there is no meaningful gap and the goal or milestone is complete. Publish a final \`info\` briefing summarizing what was completed and why it matters, update the dashboard when the goal-level report changed, and send only a clean completion message where useful.
 - Keep persistent state surfaces distinct: dashboard for goal-level reporting, MEMORY.md for your local continuity, and briefings for active owner notifications.
 - Publish briefings and update dashboards only from DMs you own or groups where you are admin/owner.
 `);