@axiomantic/garden 0.2.2 → 0.2.4

package/SKILL.md

@@ -103,6 +103,10 @@ Upon receiving the operator's responses:
 The primary conversation session acts as the Lead Orchestrator. The orchestrator directs, reviews, and weaves; it never performs large multi-file implementation edits directly when a worker fleet is active.
 </CRITICAL>
 
+<INVARIANT>
+Delegation Precedence: When instructed to "delegate", "assign", or "dispatch" work, the Lead Orchestrator MUST route tasks to active cluster workers via Rhizo (`rhizo send <worker>`, `rhizo enqueue queue:<project>:tasks`, or `rhizo task assign/claim`). Harness-internal subagents (e.g. `invoke_subagent`, `Task`, `Agent`) must ONLY be used if the operator explicitly requests a "subagent" by name, or if no cluster workers exist in the cluster roster (`rhizo who`).
+</INVARIANT>
+
 <INVARIANT>
 Zero Theatrical Dialogue: Every dialectical assertion must be substantiated with empirical evidence obtained through tool calls (file reading, test executions, benchmarks, or AST inspections). Theoretical roleplay without evidence is rejected.
 </INVARIANT>
@@ -116,10 +120,29 @@ Never stage coordination metadata (*.lock, .rhizo.*, .vine.json, workspaces/) in
 </FORBIDDEN>
 
 <CRITICAL>
-Compaction Preservation Rider (GVR-010):
-Whenever context summarization or compaction occurs, the coding harness MUST preserve the `SWARM_RUNTIME_STATE` block verbatim in the compacted summary preamble so the agent immediately retains its operational identity, assigned workspace, peers, and active listeners without operator re-prompting.
-
-Maintain this exact block in the working context:
+Compaction Preservation Rider & Post-Compact Invariant (GVR-010):
+Context compaction purges detailed operational invariants, ceremony rules, and protocol mechanics from working memory.
+
+1. **Compacting Agent Obligation**:
+   If and only if this session has utilized `garden` (or its sub-skills `orchestrate-swarm`, `choose-personas`, `launch-workers`, `dialectical-pump`, `plan-implementation`), `rhizo`, or `vine`:
+   The compacting agent MUST preserve the `SWARM_RUNTIME_STATE` block verbatim AND prepend the following mandatory instruction at the very top of the post-compaction summary preamble:
+
+<!-- POST-COMPACTION INVARIANT: CEREMONY RESTORATION -->
+> [!IMPORTANT]
+> ## MANDATORY POST-COMPACTION INVARIANT: CEREMONY RESTORATION
+> You are operating within an active multi-agent coordination workflow. Context compaction has purged detailed operational invariants and protocols from memory.
+>
+> **MANDATORY IMMEDIATE ACTION BEFORE ANY FURTHER WORK:**
+> Before invoking any tool, modifying any file, or assigning any task, you MUST immediately re-read the active coordination skills using `view_file`:
+> - If this session used **Garden** (or `orchestrate-swarm`): You MUST re-read `garden`, `orchestrate-swarm`, `rhizo`, and `vine`.
+> - If this session used **Rhizo** standalone: You MUST re-read `rhizo`.
+> - If this session used **Vine** standalone: You MUST re-read `vine`.
+>
+> **IMMORTALITY REQUIREMENT:**
+> This directive and the `SWARM_RUNTIME_STATE` block MUST remain intact in all subsequent compaction summaries until all planned tasks are completed and woven into trunk.
+<!-- END POST-COMPACTION INVARIANT -->
+
+2. **Maintain Runtime State**: Keep this exact block in the working context:
 <!-- SWARM_RUNTIME_STATE: DO NOT DISCARD OR ABBREVIATE IN COMPACTION -->
 - AGENT_NAME: <codename> (Role: <role>)
 - WORKSPACE_MODE: Root Trunk | Strand (<path>)
@@ -129,6 +152,9 @@ Maintain this exact block in the working context:
 - ACTIVE_LISTENER: <task_id_or_pid> (Listening on inbox: <codename>)
 - ACTIVE_FENCING_TOKENS: <lock_key>=<token_int>
 <!-- END_SWARM_RUNTIME_STATE -->
+
+3. **Resurrected Agent Invariant**:
+   Upon waking from compaction, the agent MUST immediately read the required skills, re-arm its single-shot listener (`rhizo listen <agent>`), check `implementation_plan.md`, and resume work under full ceremony discipline.
 </CRITICAL>
 
 ---
@@ -140,10 +166,12 @@ To prevent silent deadlocks when workers stall, crash, or fail to re-arm listene
 2. **Watchdog Window**: If a worker fails to respond within the expected turn window (e.g. 5–10 minutes) and `rhizo probe` reveals `NO_LISTENER` or unread inbox items:
    - **Escalate Immediately**: Prompt the operator via `ask_question` with the diagnostic status.
    - **Actionable Remediation**: Offer options to (1) re-arm the listener in the worker's terminal session (`rhizo listen <worker>`), (2) reboot the agent harness, or (3) reassign the task via `rhizo reroute <worker> <new_worker>`.
-3. **Orchestrator Self-Audit Watchdog & Debouncer Protocol (GVR-014)**:
-   - For harnesses supporting `schedule` (e.g. Antigravity), arm a debounced 15-minute watchdog timer (`schedule(DurationSeconds=900, Prompt="...", TimerCondition="any")`).
-   - Debouncer replaces (kills previous timer via `manage_task(Action='kill')` before arming a new one) on task dispatch, worker reports, and plan updates ("early and often"). Arriving worker traffic cancels the timer for free with 0 token overhead.
-   - When the timer fires, execute the short check: `rhizo watchdog check --agent <orchestrator> --json`. If `ACTION_REQUIRED: REARM_LISTENER`, revive `rhizo listen` in the background and debounce. When all tasks in the plan are complete (`- [x]`), stand down.
+3. **Orchestrator Self-Audit Watchdog & Stepped Backoff Protocol (GVR-014, GVR-015)**:
+   - For harnesses supporting `schedule` (e.g. Antigravity), arm a debounced watchdog timer (`schedule(DurationSeconds=cadence, Prompt="...", TimerCondition="any")`).
+   - **Stepped Backoff & 4-Strike Cap**: Starts at base 15m (900s). On consecutive quiescent checks with a stable listener, backs off (15m $\rightarrow$ 30m $\rightarrow$ 60m $\rightarrow$ 120m) and stands down at check 4 (`recommended_cadence=0`), preventing infinite token-eating polling loops.
+   - **Reset Invariant**: Resets immediately to base 15m (streak 0) on any listener failure, unread inbox backlog, outbound task dispatch (`rhizo send`/`enqueue`), worker reply, or user chat prompt.
+   - **Replace, Never Stack**: Kills previous timer via `manage_task(Action='kill')` before arming a new one. Arriving worker traffic cancels the timer automatically with zero token overhead.
+   - When the timer fires, execute the short check: `rhizo watchdog check --agent <orchestrator> --json` and follow `next_action` (`SCHEDULE_TIMER` or `STAND_DOWN`).
 
 ---
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@axiomantic/garden",
-  "version": "0.2.2",
+  "version": "0.2.4",
   "description": "Multi-Agent Swarm Orchestration, Empirical Dialectics & Ceremonies on top of Rhizo & Vine",
   "main": "bin/run.js",
   "bin": {

package/skills/garden/SKILL.md

@@ -103,6 +103,10 @@ Upon receiving the operator's responses:
 The primary conversation session acts as the Lead Orchestrator. The orchestrator directs, reviews, and weaves; it never performs large multi-file implementation edits directly when a worker fleet is active.
 </CRITICAL>
 
+<INVARIANT>
+Delegation Precedence: When instructed to "delegate", "assign", or "dispatch" work, the Lead Orchestrator MUST route tasks to active cluster workers via Rhizo (`rhizo send <worker>`, `rhizo enqueue queue:<project>:tasks`, or `rhizo task assign/claim`). Harness-internal subagents (e.g. `invoke_subagent`, `Task`, `Agent`) must ONLY be used if the operator explicitly requests a "subagent" by name, or if no cluster workers exist in the cluster roster (`rhizo who`).
+</INVARIANT>
+
 <INVARIANT>
 Zero Theatrical Dialogue: Every dialectical assertion must be substantiated with empirical evidence obtained through tool calls (file reading, test executions, benchmarks, or AST inspections). Theoretical roleplay without evidence is rejected.
 </INVARIANT>
@@ -116,10 +120,29 @@ Never stage coordination metadata (*.lock, .rhizo.*, .vine.json, workspaces/) in
 </FORBIDDEN>
 
 <CRITICAL>
-Compaction Preservation Rider (GVR-010):
-Whenever context summarization or compaction occurs, the coding harness MUST preserve the `SWARM_RUNTIME_STATE` block verbatim in the compacted summary preamble so the agent immediately retains its operational identity, assigned workspace, peers, and active listeners without operator re-prompting.
-
-Maintain this exact block in the working context:
+Compaction Preservation Rider & Post-Compact Invariant (GVR-010):
+Context compaction purges detailed operational invariants, ceremony rules, and protocol mechanics from working memory.
+
+1. **Compacting Agent Obligation**:
+   If and only if this session has utilized `garden` (or its sub-skills `orchestrate-swarm`, `choose-personas`, `launch-workers`, `dialectical-pump`, `plan-implementation`), `rhizo`, or `vine`:
+   The compacting agent MUST preserve the `SWARM_RUNTIME_STATE` block verbatim AND prepend the following mandatory instruction at the very top of the post-compaction summary preamble:
+
+<!-- POST-COMPACTION INVARIANT: CEREMONY RESTORATION -->
+> [!IMPORTANT]
+> ## MANDATORY POST-COMPACTION INVARIANT: CEREMONY RESTORATION
+> You are operating within an active multi-agent coordination workflow. Context compaction has purged detailed operational invariants and protocols from memory.
+>
+> **MANDATORY IMMEDIATE ACTION BEFORE ANY FURTHER WORK:**
+> Before invoking any tool, modifying any file, or assigning any task, you MUST immediately re-read the active coordination skills using `view_file`:
+> - If this session used **Garden** (or `orchestrate-swarm`): You MUST re-read `garden`, `orchestrate-swarm`, `rhizo`, and `vine`.
+> - If this session used **Rhizo** standalone: You MUST re-read `rhizo`.
+> - If this session used **Vine** standalone: You MUST re-read `vine`.
+>
+> **IMMORTALITY REQUIREMENT:**
+> This directive and the `SWARM_RUNTIME_STATE` block MUST remain intact in all subsequent compaction summaries until all planned tasks are completed and woven into trunk.
+<!-- END POST-COMPACTION INVARIANT -->
+
+2. **Maintain Runtime State**: Keep this exact block in the working context:
 <!-- SWARM_RUNTIME_STATE: DO NOT DISCARD OR ABBREVIATE IN COMPACTION -->
 - AGENT_NAME: <codename> (Role: <role>)
 - WORKSPACE_MODE: Root Trunk | Strand (<path>)
@@ -129,6 +152,9 @@ Maintain this exact block in the working context:
 - ACTIVE_LISTENER: <task_id_or_pid> (Listening on inbox: <codename>)
 - ACTIVE_FENCING_TOKENS: <lock_key>=<token_int>
 <!-- END_SWARM_RUNTIME_STATE -->
+
+3. **Resurrected Agent Invariant**:
+   Upon waking from compaction, the agent MUST immediately read the required skills, re-arm its single-shot listener (`rhizo listen <agent>`), check `implementation_plan.md`, and resume work under full ceremony discipline.
 </CRITICAL>
 
 ---
@@ -140,10 +166,12 @@ To prevent silent deadlocks when workers stall, crash, or fail to re-arm listene
 2. **Watchdog Window**: If a worker fails to respond within the expected turn window (e.g. 5–10 minutes) and `rhizo probe` reveals `NO_LISTENER` or unread inbox items:
    - **Escalate Immediately**: Prompt the operator via `ask_question` with the diagnostic status.
    - **Actionable Remediation**: Offer options to (1) re-arm the listener in the worker's terminal session (`rhizo listen <worker>`), (2) reboot the agent harness, or (3) reassign the task via `rhizo reroute <worker> <new_worker>`.
-3. **Orchestrator Self-Audit Watchdog & Debouncer Protocol (GVR-014)**:
-   - For harnesses supporting `schedule` (e.g. Antigravity), arm a debounced 15-minute watchdog timer (`schedule(DurationSeconds=900, Prompt="...", TimerCondition="any")`).
-   - Debouncer replaces (kills previous timer via `manage_task(Action='kill')` before arming a new one) on task dispatch, worker reports, and plan updates ("early and often"). Arriving worker traffic cancels the timer for free with 0 token overhead.
-   - When the timer fires, execute the short check: `rhizo watchdog check --agent <orchestrator> --json`. If `ACTION_REQUIRED: REARM_LISTENER`, revive `rhizo listen` in the background and debounce. When all tasks in the plan are complete (`- [x]`), stand down.
+3. **Orchestrator Self-Audit Watchdog & Stepped Backoff Protocol (GVR-014, GVR-015)**:
+   - For harnesses supporting `schedule` (e.g. Antigravity), arm a debounced watchdog timer (`schedule(DurationSeconds=cadence, Prompt="...", TimerCondition="any")`).
+   - **Stepped Backoff & 4-Strike Cap**: Starts at base 15m (900s). On consecutive quiescent checks with a stable listener, backs off (15m $\rightarrow$ 30m $\rightarrow$ 60m $\rightarrow$ 120m) and stands down at check 4 (`recommended_cadence=0`), preventing infinite token-eating polling loops.
+   - **Reset Invariant**: Resets immediately to base 15m (streak 0) on any listener failure, unread inbox backlog, outbound task dispatch (`rhizo send`/`enqueue`), worker reply, or user chat prompt.
+   - **Replace, Never Stack**: Kills previous timer via `manage_task(Action='kill')` before arming a new one. Arriving worker traffic cancels the timer automatically with zero token overhead.
+   - When the timer fires, execute the short check: `rhizo watchdog check --agent <orchestrator> --json` and follow `next_action` (`SCHEDULE_TIMER` or `STAND_DOWN`).
 
 ---
 

package/skills/orchestrate-swarm/SKILL.md

@@ -18,10 +18,35 @@ The main chat session assumes the role of **Lead Orchestrator**:
 - **Gated Integration**: Never run `git merge` directly. Only weave branches that have passed both Key 1 (mechanical merge-tree) and Key 2 (live compiler/tests) inside their Vine strands.
 
 <CRITICAL>
-Compaction Preservation Rider (GVR-010):
-Whenever context summarization or compaction occurs, the coding harness MUST preserve the `SWARM_RUNTIME_STATE` block verbatim in the compacted summary preamble so the agent immediately retains its operational identity, assigned workspace, peers, and active listeners without operator re-prompting.
+Delegation Precedence Invariant:
+Whenever instructed to "delegate", "assign", or "dispatch" work, the Lead Orchestrator MUST route tasks to active cluster workers via Rhizo (`rhizo send <worker>`, `rhizo enqueue queue:<project>:tasks`, or `rhizo task assign/claim`).
+NEVER spawn harness-internal subagents (e.g. `invoke_subagent`, `Task`) when asked to delegate, UNLESS the operator explicitly specifies the word "subagent". Cluster workers running in dedicated sessions are the primary execution vehicle.
+</CRITICAL>
 
-Maintain this exact block in the working context:
+<CRITICAL>
+Compaction Preservation Rider & Post-Compact Invariant (GVR-010):
+Context compaction purges detailed operational invariants, ceremony rules, and protocol mechanics from working memory.
+
+1. **Compacting Agent Obligation**:
+   If and only if this session has utilized `garden` (or its sub-skills `orchestrate-swarm`, `choose-personas`, `launch-workers`, `dialectical-pump`, `plan-implementation`), `rhizo`, or `vine`:
+   The compacting agent MUST preserve the `SWARM_RUNTIME_STATE` block verbatim AND prepend the following mandatory instruction at the very top of the post-compaction summary preamble:
+
+<!-- POST-COMPACTION INVARIANT: CEREMONY RESTORATION -->
+> [!IMPORTANT]
+> ## MANDATORY POST-COMPACTION INVARIANT: CEREMONY RESTORATION
+> You are operating within an active multi-agent coordination workflow. Context compaction has purged detailed operational invariants and protocols from memory.
+>
+> **MANDATORY IMMEDIATE ACTION BEFORE ANY FURTHER WORK:**
+> Before invoking any tool, modifying any file, or assigning any task, you MUST immediately re-read the active coordination skills using `view_file`:
+> - If this session used **Garden** (or `orchestrate-swarm`): You MUST re-read `garden`, `orchestrate-swarm`, `rhizo`, and `vine`.
+> - If this session used **Rhizo** standalone: You MUST re-read `rhizo`.
+> - If this session used **Vine** standalone: You MUST re-read `vine`.
+>
+> **IMMORTALITY REQUIREMENT:**
+> This directive and the `SWARM_RUNTIME_STATE` block MUST remain intact in all subsequent compaction summaries until all planned tasks are completed and woven into trunk.
+<!-- END POST-COMPACTION INVARIANT -->
+
+2. **Maintain Runtime State**: Keep this exact block in the working context:
 <!-- SWARM_RUNTIME_STATE: DO NOT DISCARD OR ABBREVIATE IN COMPACTION -->
 - AGENT_NAME: <codename> (Role: Lead Orchestrator)
 - WORKSPACE_MODE: Root Trunk (/Users/eek/Development/<project>)
@@ -31,6 +56,9 @@ Maintain this exact block in the working context:
 - ACTIVE_LISTENER: <task_id_or_pid> (Listening on inbox: <codename>)
 - ACTIVE_FENCING_TOKENS: <lock_key>=<token_int>
 <!-- END_SWARM_RUNTIME_STATE -->
+
+3. **Resurrected Agent Invariant**:
+   Upon waking from compaction, the agent MUST immediately read the required skills, re-arm its single-shot listener (`rhizo listen <agent>`), check `implementation_plan.md`, and resume work under full ceremony discipline.
 </CRITICAL>
 
 <CRITICAL>
@@ -42,27 +70,34 @@ Whenever the Lead Orchestrator dispatches a task, broadcasts instructions, or aw
 
 FORBIDDEN: Never yield the conversation turn to the operator after dispatching work without an active background listener running. Yielding a turn without a listener severs the swarm's physical lifeline, trapping worker replies in Redis and causing silent swarm stalls.
 
-Safety Net (Scheduled Timer Watchdog & Debouncer Protocol):
+Safety Net (Scheduled Timer Watchdog & Debouncer Protocol — Stepped Backoff & 4-Strike Cap):
 In harnesses supporting `schedule` (e.g. Google Antigravity), arm a debounced watchdog timer to ensure an orchestrator session is never abandoned if a listener fails to arm or terminates prematurely.
-- **Cadence**: 15 minutes (`DurationSeconds=900`, range 10m–30m / 600s–1800s). Defaulting to 15m avoids slurping token budgets while guaranteeing a 15m upper bound on any stall.
+- **Base Cadence & Stepped Backoff**:
+  - Initial / After Activity: Base 15 minutes (`DurationSeconds=900`).
+  - Quiescent Check 1 (Streak 1): 30 minutes (`DurationSeconds=1800`).
+  - Quiescent Check 2 (Streak 2): 60 minutes (`DurationSeconds=3600`).
+  - Quiescent Check 3 (Streak 3): 120 minutes (`DurationSeconds=7200`).
+  - Quiescent Check 4 (Streak 4): **Stand Down** (`recommended_cadence=0`, do not reschedule).
+- **The Non-Exponential Reset Invariant**:
+  The quiescent streak and timer cadence IMMEDIATELY reset to 0 (base 15m / 900s) upon:
+  1. Any listener failure or missing process (`ACTION_REQUIRED: REARM_LISTENER`).
+  2. Any unread inbox backlog (`ACTION_REQUIRED: UNREAD_MESSAGES`).
+  3. Any outbound task dispatch (`rhizo send`, `rhizo enqueue`, `rhizo reply`).
+  4. Any worker gate report or message receipt.
+  5. Any operator interaction or new prompt in chat.
 - **Replace, Never Stack Invariant**:
   Harnesses prohibit concurrent timers with `TimerCondition="any"`. Before setting a timer, inspect running tasks with `manage_task(Action='list')`. If an existing watchdog task is active (`toolName == "schedule"` or prompt includes `[RHIZO WATCHDOG]`), cancel it via `manage_task(Action='kill', TaskId=...)`.
-- **Debounce Triggers (Early and Often)**:
-  Run the debouncer subroutine on:
-  1. Task Dispatch (`rhizo send`, `rhizo enqueue`).
-  2. Worker Message / Gate Report receipt.
-  3. Implementation Plan updates (`implementation_plan.md` checkboxes).
-  4. Watchdog Wakeup turn (if tasks are still in flight).
-- **Stand Down Invariant**:
-  When all tasks in `implementation_plan.md` are complete (`- [x]` 100%), kill any running watchdog timer and do not reschedule.
+- **Stand Down Invariants**:
+  1. When all tasks in `implementation_plan.md` are complete (`- [x]` 100%), kill any running watchdog timer and do not reschedule.
+  2. When the watchdog reaches `substatus: "MAX_STREAK_REACHED"` (streak 4/4), stand down and do not reschedule. The background listener process (`rhizo listen`) remains continuously active on Redis `BRPOP` and will wake the session on any new message.
 - **Zero-Token Happy Path**:
-  Because `TimerCondition="any"` is set, any arriving worker message or background task completion automatically cancels the timer early before it expires. The timer only fires if the orchestrator was silent and deaf for a full 15 minutes.
+  Because `TimerCondition="any"` is set, any arriving worker message or background task completion automatically cancels the timer early before it expires. The timer only fires if the orchestrator was silent and deaf for the full cadence duration.
 - **The Short Check (When Timer Fires)**:
   Run `rhizo watchdog check --agent <name> --json`.
-  * If `ACTION_REQUIRED: REARM_LISTENER`: start `rhizo listen <name>` in background and debounce timer.
-  * If `ACTION_REQUIRED: UNREAD_MESSAGES`: drain messages with `rhizo drain 10 <name>`, start listener, and debounce.
-  * If `OK: LISTENING`: listener is healthy; debounce timer and return to sleep.
-  * If `STAND_DOWN: IDLE`: no tasks in flight; stand down.
+  * If `ACTION_REQUIRED: REARM_LISTENER`: start `rhizo listen <name>` in background and schedule base timer (`DurationSeconds=900`).
+  * If `ACTION_REQUIRED: UNREAD_MESSAGES`: drain messages with `rhizo drain 10 <name>`, start listener, and schedule base timer (`DurationSeconds=900`).
+  * If `OK: LISTENING`: listener is healthy; check `recommended_cadence`. If `next_action == "SCHEDULE_TIMER"`, schedule timer with `DurationSeconds=recommended_cadence` and return to sleep with minimal or no user-facing chat output.
+  * If `STAND_DOWN: MAX_STREAK_REACHED` or `STAND_DOWN: IDLE`: stand down and do NOT reschedule.
 </CRITICAL>
 
 ---