diff --git a/REPORT.md b/REPORT.md
index 2ca94f0..fdf4686 100644
--- a/REPORT.md
+++ b/REPORT.md
@@ -20,13 +20,9 @@ Commander parses `--no-snapshot` as `{ snapshot: false }`. Verified: `opts.snaps
 
 ## High
 
-### 4. SKILL.md parser fragile — OPEN
+### 4. SKILL.md parser fragile — FIXED
 **File:** `agent/agent.py:69-138`
-- Requires exact 2-space/4-space indentation
-- No error reporting when skills fail to load
-- No validation of parameter types
-- Case-sensitive boolean parsing (`true` works, `True` doesn't)
-- Assumes Unix line endings
+**Fix:** Added error logging on parse failures, flexible indent detection (2+ spaces), CRLF normalization, boolean case-insensitive (`true`/`True`/`yes`/`1`), parameter type validation with warnings.
 
 ### 5. SSH host key verification disabled — DOCUMENTED
 **Files:** `src/ssh.ts:104`, `src/agent-manager.ts`, `src/overseer.ts`
@@ -44,22 +40,22 @@ After save_memory, only MEMORY.md index was reloaded. Individual memory files we
 Same SSH options array repeated 4+ times.
 **Fix:** Extracted `SSH_OPTS` constant in both files.
 
-### 8. Process termination inconsistent — OPEN
+### 8. Process termination inconsistent — OPEN (low risk)
 **Files:** `src/firecracker-vm.ts:140-164`, `src/agent-manager.ts:319-332`
-Two different implementations: one uses SIGTERM→wait→SIGKILL on ChildProcess, the other uses `process.kill(pid, 0)` polling. Should consolidate.
+Two different implementations. The ChildProcess version uses SIGTERM→SIGKILL, the PID version uses polling. Both work, different contexts (owned vs adopted processes).
 
-### 9. killall python3 hardcoded — OPEN
-**Files:** `src/agent-manager.ts:310`, `src/agent-manager.ts:430`
-Assumes agent is always a python3 process. Fragile if agent wrapping changes.
+### 9. killall python3 hardcoded — FIXED
+**Files:** `src/agent-manager.ts`
+**Fix:** Replaced `killall python3` with `pkill -f 'agent.py'`. Targets the specific script, not all python3 processes.
 
 ### 10. Test suite expects researcher template — FIXED
 **File:** `tests/test-suite.sh:105`
 Test asserts `researcher` template exists, but install script only created worker, coder, quick.
 **Fix:** Added researcher and creative templates to `scripts/install.sh`.
 
-### 11. Bare exception handlers — OPEN
-**File:** `src/agent-manager.ts` — 8+ `catch {}` blocks
-Swallow all errors silently. Makes debugging impossible.
+### 11. Bare exception handlers — FIXED
+**File:** `src/agent-manager.ts`
+**Fix:** Added error logging to cleanup catch blocks in listAgents and reconcileAgents. Remaining bare catches are intentional best-effort cleanup (umount, rmdir, unlink).
 
 ### 12. agent.py monolithic (598 lines) — OPEN
 **File:** `agent/agent.py`
@@ -83,17 +79,19 @@ Left over after switching to `atomicWritePool`. Removed.
 ### 17. Ollama response size unbounded
 **File:** `agent/agent.py:338` — `resp.read()` with no size limit.
 
-### 18. IRC message splitting at 400 chars
-**File:** `agent/agent.py:266` — IRC limit is 512 bytes including protocol overhead. Safe limit is ~380 chars.
+### 18. IRC message splitting at 400 chars — FIXED
+**File:** `agent/agent.py:266`
+**Fix:** Reduced to 380 chars to stay within IRC's 512-byte line limit.
 
-### 19. Thread safety on _last_response_time
-**File:** `agent/agent.py:485` — global updated without lock, two rapid requests could both pass cooldown.
+### 19. Thread safety on _last_response_time — FIXED
+**File:** `agent/agent.py:485`
+**Fix:** Added `_cooldown_lock` (threading.Lock) around cooldown check-and-set.
 
 ## Summary
 
 | Status | Count |
 |---|---|
-| Fixed | 7 |
+| Fixed | 13 |
 | False alarm | 1 |
-| Open (medium+) | 6 |
-| Open (low) | 6 |
+| Open (medium) | 2 |
+| Open (low) | 4 |