chore(agents): fix ca-test-infra-improver health comment spam

Approved proposal: #3385
Pattern: workflow_fix
Evidence: Agent posted 44 of 52 comments in 20 minutes, drowning out all
other agent signals. The cycle % 60 guard was ambiguous and ignored by the
model.
Fix: Replace fragile modulo guard with explicit timestamp-based rate
limiting (10-minute minimum between health posts). Add prominent CRITICAL
warning about rate limiting. Clarify that health posting runs in the outer
loop, not the inner monitoring loop.

ISSUES CLOSED: #3385

.opencode/agents/ca-test-infra-improver.md

@@ -133,6 +133,14 @@ If no spec context is provided, invoke `ca-ref-reader` once at startup.
 
 ### Pool Supervision Loop
 
+**CRITICAL: Health Comment Rate Limiting.** Do NOT post health comments on
+every monitoring iteration. Health comments MUST be posted **at most once
+every 10 minutes**, enforced by a `last_health_post_time` timestamp. If
+less than 10 minutes have elapsed since the last health comment, do NOT
+post. The ONLY exception is a **significant state change** (worker
+completed, all areas analyzed, new worker dispatched after a slot freed
+up). Even state-change posts are limited to at most one per 60 seconds.
+
 ```
 # Check if session state issue number was provided
 if SESSION_STATE_ISSUE_NUMBER not provided:
@@ -157,6 +165,8 @@ analysis_areas = [
 analyzed_areas = set()
 findings_total = 0
 cycle = 0
+last_health_post_time = 0       # Timestamp of last health comment (epoch seconds)
+HEALTH_INTERVAL_SECONDS = 600   # 10 minutes between health comments
 
 # ── RESUME: Adopt existing worker sessions from previous run ─────
 EXISTING_WORKERS = bash("curl -s ${SERVER}/session | python3 -c \"
@@ -233,8 +243,12 @@ LOOP:
                     NEW_SID = create session + prompt_async for next_area
                     active[next_area] = NEW_SID
 
-    # ── Post progress ────────────────────────────────────────────
-    if cycle % 60 == 0:  # Every ~10 minutes with 10-second monitoring
+    # ── Post progress (RATE LIMITED — see rules above) ───────────
+    # IMPORTANT: This section runs ONCE per outer supervision cycle,
+    # NOT inside the inner "while active" monitoring loop above.
+    # Only post if at least HEALTH_INTERVAL_SECONDS have elapsed.
+    current_time = time.time()  # or equivalent epoch seconds
+    if current_time - last_health_post_time >= HEALTH_INTERVAL_SECONDS:
         post comment on session state issue:
             "[HEALTH] ca-test-infra-improver | Iteration: <cycle> | Status: active\n" +
              "- Type: pool-supervisor\n" +
@@ -246,6 +260,7 @@ LOOP:
              "---\n" +
              "**Automated by CleverAgents Bot**\n" +
              "Supervisor: Test Infrastructure | Agent: ca-test-infra-improver"
+        last_health_post_time = current_time
 ```
 
 ---