diff --git a/.agents/skills/gstack-retro/SKILL.md b/.agents/skills/gstack-retro/SKILL.md
index a0b796ba..a56cc3b1 100644
--- a/.agents/skills/gstack-retro/SKILL.md
+++ b/.agents/skills/gstack-retro/SKILL.md
@@ -329,8 +329,11 @@ git log origin/<default> --since="<window>" --oneline --grep="test(qa):" --grep=
 # 12. gstack skill usage telemetry (if available)
 cat ~/.gstack/analytics/skill-usage.jsonl 2>/dev/null || true
 
-# 12. Test files changed in window
+# 13. Test files changed in window
 git log origin/<default> --since="<window>" --format="" --name-only | grep -E '\.(test|spec)\.' | sort -u | wc -l
+
+# 14. Ship run log (if available)
+~/.codex/skills/gstack/bin/gstack-ship-log read --window <window> 2>/dev/null || true
 ```
 
 ### Step 2: Compute Metrics
@@ -403,6 +406,21 @@ If moments exist, list them:
 
 If the JSONL file doesn't exist or has no entries in the window, skip the Eureka Moments row.
 
+**Ship Velocity (if ship log exists):** Read the ship log output from Step 1, command 14. If entries exist within the retro window, aggregate:
+- Total `/ship` runs
+- PRs created (count entries with non-empty `pr_url`)
+- Average review findings per ship run (`review_findings`)
+- Total Greptile catches vs false positives (`greptile_fixed` vs `greptile_fps`)
+- TODOs completed via `/ship` (sum of `todos_completed`)
+- Test coverage delta (average `coverage_after - coverage_before`)
+
+Include in the metrics table:
+```
+| Ship Velocity | N /ship runs · M PRs · avg X review findings · Y Greptile catches |
+```
+
+If the ship log doesn't exist or has no entries in the window (`NO_SHIP_LOG` returned), skip the Ship Velocity row.
+
 ### Step 3: Commit Time Distribution
 
 Show hourly histogram in local time using bar chart:
@@ -598,7 +616,7 @@ Use the Write tool to save the JSON file with this schema:
 }
 ```
 
-**Note:** Only include the `greptile` field if `~/.gstack/greptile-history.md` exists and has entries within the time window. Only include the `backlog` field if `TODOS.md` exists. Only include the `test_health` field if test files were found (command 10 returns > 0). If any has no data, omit the field entirely.
+**Note:** Only include the `greptile` field if `~/.gstack/greptile-history.md` exists and has entries within the time window. Only include the `backlog` field if `TODOS.md` exists. Only include the `test_health` field if test files were found (command 10 returns > 0). Only include `ship_velocity` if the ship log has entries in the window. If any has no data, omit the field entirely.
 
 Include test health data in the JSON when test files exist:
 ```json
@@ -621,6 +639,19 @@ Include backlog data in the JSON when TODOS.md exists:
   }
 ```
 
+Include ship velocity data in the JSON when the ship log has entries in the window:
+```json
+  "ship_velocity": {
+    "ship_runs": 5,
+    "prs_created": 5,
+    "avg_review_findings": 2.4,
+    "greptile_catches": 3,
+    "greptile_fps": 1,
+    "todos_completed": 8,
+    "avg_coverage_delta": 3
+  }
+```
+
 ### Step 14: Write the Narrative
 
 Structure the output as:
@@ -650,13 +681,14 @@ Narrative interpreting what the team-wide patterns mean:
 - Notable patterns: do team members code at the same time or in shifts?
 
 ### Shipping Velocity
-(from Steps 5-7)
+(from Steps 5-7 + ship log)
 
 Narrative covering:
 - Commit type mix and what it reveals
 - PR size distribution and what it reveals about shipping cadence
 - Fix-chain detection (sequences of fix commits on the same subsystem)
 - Version bump discipline
+- Ship log trends (if data exists): `/ship` runs per period, review finding rates, Greptile signal over time, test coverage growth from `/ship` auto-generation
 
 ### Code Quality Signals
 - Test LOC ratio trend
diff --git a/.agents/skills/gstack-ship/SKILL.md b/.agents/skills/gstack-ship/SKILL.md
index 442c4a72..c438a359 100644
--- a/.agents/skills/gstack-ship/SKILL.md
+++ b/.agents/skills/gstack-ship/SKILL.md
@@ -1068,7 +1068,34 @@ EOF
 )"
 ```
 
-**Output the PR URL** — then proceed to Step 8.5.
+**Output the PR URL** — then proceed to Step 8.25.
+
+---
+
+## Step 8.25: Ship Log
+
+Append a structured entry to the ship log for `/retro` velocity tracking. This step is fully automatic and must never fail the workflow.
+
+Collect the data from earlier steps and write a single JSONL entry:
+
+```bash
+source <(~/.codex/skills/gstack/bin/gstack-slug 2>/dev/null)
+~/.codex/skills/gstack/bin/gstack-ship-log '{"ts":"'"$(date -u +%Y-%m-%dT%H:%M:%SZ)"'","version":"VERSION","branch":"'"$BRANCH"'","repo":"'"$SLUG"'","pr_url":"PR_URL","review_findings":N,"review_auto_fixed":N,"greptile_comments":N,"greptile_fixed":N,"greptile_fps":N,"todos_completed":N,"tests_passed":true,"coverage_before":N,"coverage_after":N}'
+```
+
+Substitute the capitalized placeholders from data gathered in earlier steps:
+- `VERSION`: the new version from Step 4
+- `PR_URL`: the PR URL output from Step 8
+- `review_findings`: total issues found in Step 3.5 (0 if none)
+- `review_auto_fixed`: count of auto-fixed issues in Step 3.5
+- `greptile_comments`: total Greptile comments in Step 3.75 (0 if skipped)
+- `greptile_fixed`: Greptile comments fixed in Step 3.75
+- `greptile_fps`: Greptile false positives in Step 3.75
+- `todos_completed`: items marked complete in Step 5.5 (0 if none)
+- `tests_passed`: `true` (always true here — Step 3 would have stopped if tests failed)
+- `coverage_before`/`coverage_after`: test file counts from Step 3.4 (0 if skipped)
+
+If the `gstack-ship-log` script is not found or the write fails, warn once and continue. Never stop the ship workflow for a log failure.
 
 ---
 
diff --git a/bin/gstack-ship-log b/bin/gstack-ship-log
new file mode 100755
index 00000000..741d3267
--- /dev/null
+++ b/bin/gstack-ship-log
@@ -0,0 +1,71 @@
+#!/usr/bin/env bash
+# gstack-ship-log — append a structured JSON entry to the ship log
+#
+# Usage:
+#   gstack-ship-log '{"ts":"...","version":"...","branch":"...","pr_url":"..."}'
+#   gstack-ship-log read              — print ship log contents (or NO_SHIP_LOG)
+#   gstack-ship-log read --window 7d  — print entries from last N days
+#
+# The ship log lives at ~/.gstack/analytics/ship-log.jsonl.
+# /ship appends an entry after each PR creation.
+# /retro reads the log for shipping velocity metrics.
+#
+# Env overrides (for testing):
+#   GSTACK_STATE_DIR  — override ~/.gstack state directory
+set -uo pipefail
+
+STATE_DIR="${GSTACK_STATE_DIR:-$HOME/.gstack}"
+LOG_FILE="$STATE_DIR/analytics/ship-log.jsonl"
+
+case "${1:-}" in
+  read)
+    if [ ! -f "$LOG_FILE" ]; then
+      echo "NO_SHIP_LOG"
+      exit 0
+    fi
+    # Optional --window filter
+    if [ "${2:-}" = "--window" ] && [ -n "${3:-}" ]; then
+      WINDOW="${3}"
+      # Extract numeric value and unit
+      DAYS=0
+      case "$WINDOW" in
+        *d) DAYS="${WINDOW%d}" ;;
+        *h) DAYS=0 ;;  # sub-day: include everything (awk filters by timestamp)
+        *)  DAYS=7 ;;
+      esac
+      if [ "$DAYS" -gt 0 ] 2>/dev/null; then
+        if date -v-1d +%Y-%m-%d >/dev/null 2>&1; then
+          CUTOFF="$(date -v-${DAYS}d -u +%Y-%m-%dT%H:%M:%SZ)"
+        else
+          CUTOFF="$(date -u -d "$DAYS days ago" +%Y-%m-%dT%H:%M:%SZ 2>/dev/null || echo "2000-01-01T00:00:00Z")"
+        fi
+        awk -F'"' -v cutoff="$CUTOFF" '
+          /"ts":"/ {
+            for (i=1; i<=NF; i++) {
+              if ($i == "ts" && $(i+1) ~ /^:/) {
+                if ($(i+2) >= cutoff) { print; break }
+              }
+            }
+          }
+        ' "$LOG_FILE"
+      else
+        cat "$LOG_FILE"
+      fi
+    else
+      cat "$LOG_FILE"
+    fi
+    ;;
+  -h|--help)
+    sed -n '2,/^[^#]/{ /^#/s/^# \{0,1\}//p; }' "$0"
+    exit 0
+    ;;
+  "")
+    echo "Usage: gstack-ship-log '{json}' or gstack-ship-log read" >&2
+    exit 1
+    ;;
+  *)
+    # Append mode: $1 is a JSON string
+    mkdir -p "$STATE_DIR/analytics"
+    echo "$1" >> "$LOG_FILE"
+    ;;
+esac
diff --git a/retro/SKILL.md b/retro/SKILL.md
index 635b5747..338f8684 100644
--- a/retro/SKILL.md
+++ b/retro/SKILL.md
@@ -336,8 +336,11 @@ git log origin/<default> --since="<window>" --oneline --grep="test(qa):" --grep=
 # 12. gstack skill usage telemetry (if available)
 cat ~/.gstack/analytics/skill-usage.jsonl 2>/dev/null || true
 
-# 12. Test files changed in window
+# 13. Test files changed in window
 git log origin/<default> --since="<window>" --format="" --name-only | grep -E '\.(test|spec)\.' | sort -u | wc -l
+
+# 14. Ship run log (if available)
+~/.claude/skills/gstack/bin/gstack-ship-log read --window <window> 2>/dev/null || true
 ```
 
 ### Step 2: Compute Metrics
@@ -410,6 +413,21 @@ If moments exist, list them:
 
 If the JSONL file doesn't exist or has no entries in the window, skip the Eureka Moments row.
 
+**Ship Velocity (if ship log exists):** Read the ship log output from Step 1, command 14. If entries exist within the retro window, aggregate:
+- Total `/ship` runs
+- PRs created (count entries with non-empty `pr_url`)
+- Average review findings per ship run (`review_findings`)
+- Total Greptile catches vs false positives (`greptile_fixed` vs `greptile_fps`)
+- TODOs completed via `/ship` (sum of `todos_completed`)
+- Test coverage delta (average `coverage_after - coverage_before`)
+
+Include in the metrics table:
+```
+| Ship Velocity | N /ship runs · M PRs · avg X review findings · Y Greptile catches |
+```
+
+If the ship log doesn't exist or has no entries in the window (`NO_SHIP_LOG` returned), skip the Ship Velocity row.
+
 ### Step 3: Commit Time Distribution
 
 Show hourly histogram in local time using bar chart:
@@ -605,7 +623,7 @@ Use the Write tool to save the JSON file with this schema:
 }
 ```
 
-**Note:** Only include the `greptile` field if `~/.gstack/greptile-history.md` exists and has entries within the time window. Only include the `backlog` field if `TODOS.md` exists. Only include the `test_health` field if test files were found (command 10 returns > 0). If any has no data, omit the field entirely.
+**Note:** Only include the `greptile` field if `~/.gstack/greptile-history.md` exists and has entries within the time window. Only include the `backlog` field if `TODOS.md` exists. Only include the `test_health` field if test files were found (command 10 returns > 0). Only include `ship_velocity` if the ship log has entries in the window. If any has no data, omit the field entirely.
 
 Include test health data in the JSON when test files exist:
 ```json
@@ -628,6 +646,19 @@ Include backlog data in the JSON when TODOS.md exists:
   }
 ```
 
+Include ship velocity data in the JSON when the ship log has entries in the window:
+```json
+  "ship_velocity": {
+    "ship_runs": 5,
+    "prs_created": 5,
+    "avg_review_findings": 2.4,
+    "greptile_catches": 3,
+    "greptile_fps": 1,
+    "todos_completed": 8,
+    "avg_coverage_delta": 3
+  }
+```
+
 ### Step 14: Write the Narrative
 
 Structure the output as:
@@ -657,13 +688,14 @@ Narrative interpreting what the team-wide patterns mean:
 - Notable patterns: do team members code at the same time or in shifts?
 
 ### Shipping Velocity
-(from Steps 5-7)
+(from Steps 5-7 + ship log)
 
 Narrative covering:
 - Commit type mix and what it reveals
 - PR size distribution and what it reveals about shipping cadence
 - Fix-chain detection (sequences of fix commits on the same subsystem)
 - Version bump discipline
+- Ship log trends (if data exists): `/ship` runs per period, review finding rates, Greptile signal over time, test coverage growth from `/ship` auto-generation
 
 ### Code Quality Signals
 - Test LOC ratio trend
diff --git a/retro/SKILL.md.tmpl b/retro/SKILL.md.tmpl
index b3fe8046..5caa938e 100644
--- a/retro/SKILL.md.tmpl
+++ b/retro/SKILL.md.tmpl
@@ -112,8 +112,11 @@ git log origin/<default> --since="<window>" --oneline --grep="test(qa):" --grep=
 # 12. gstack skill usage telemetry (if available)
 cat ~/.gstack/analytics/skill-usage.jsonl 2>/dev/null || true
 
-# 12. Test files changed in window
+# 13. Test files changed in window
 git log origin/<default> --since="<window>" --format="" --name-only | grep -E '\.(test|spec)\.' | sort -u | wc -l
+
+# 14. Ship run log (if available)
+~/.claude/skills/gstack/bin/gstack-ship-log read --window <window> 2>/dev/null || true
 ```
 
 ### Step 2: Compute Metrics
@@ -186,6 +189,21 @@ If moments exist, list them:
 
 If the JSONL file doesn't exist or has no entries in the window, skip the Eureka Moments row.
 
+**Ship Velocity (if ship log exists):** Read the ship log output from Step 1, command 14. If entries exist within the retro window, aggregate:
+- Total `/ship` runs
+- PRs created (count entries with non-empty `pr_url`)
+- Average review findings per ship run (`review_findings`)
+- Total Greptile catches vs false positives (`greptile_fixed` vs `greptile_fps`)
+- TODOs completed via `/ship` (sum of `todos_completed`)
+- Test coverage delta (average `coverage_after - coverage_before`)
+
+Include in the metrics table:
+```
+| Ship Velocity | N /ship runs · M PRs · avg X review findings · Y Greptile catches |
+```
+
+If the ship log doesn't exist or has no entries in the window (`NO_SHIP_LOG` returned), skip the Ship Velocity row.
+
 ### Step 3: Commit Time Distribution
 
 Show hourly histogram in local time using bar chart:
@@ -381,7 +399,7 @@ Use the Write tool to save the JSON file with this schema:
 }
 ```
 
-**Note:** Only include the `greptile` field if `~/.gstack/greptile-history.md` exists and has entries within the time window. Only include the `backlog` field if `TODOS.md` exists. Only include the `test_health` field if test files were found (command 10 returns > 0). If any has no data, omit the field entirely.
+**Note:** Only include the `greptile` field if `~/.gstack/greptile-history.md` exists and has entries within the time window. Only include the `backlog` field if `TODOS.md` exists. Only include the `test_health` field if test files were found (command 10 returns > 0). Only include `ship_velocity` if the ship log has entries in the window. If any has no data, omit the field entirely.
 
 Include test health data in the JSON when test files exist:
 ```json
@@ -404,6 +422,19 @@ Include backlog data in the JSON when TODOS.md exists:
   }
 ```
 
+Include ship velocity data in the JSON when the ship log has entries in the window:
+```json
+  "ship_velocity": {
+    "ship_runs": 5,
+    "prs_created": 5,
+    "avg_review_findings": 2.4,
+    "greptile_catches": 3,
+    "greptile_fps": 1,
+    "todos_completed": 8,
+    "avg_coverage_delta": 3
+  }
+```
+
 ### Step 14: Write the Narrative
 
 Structure the output as:
@@ -433,13 +464,14 @@ Narrative interpreting what the team-wide patterns mean:
 - Notable patterns: do team members code at the same time or in shifts?
 
 ### Shipping Velocity
-(from Steps 5-7)
+(from Steps 5-7 + ship log)
 
 Narrative covering:
 - Commit type mix and what it reveals
 - PR size distribution and what it reveals about shipping cadence
 - Fix-chain detection (sequences of fix commits on the same subsystem)
 - Version bump discipline
+- Ship log trends (if data exists): `/ship` runs per period, review finding rates, Greptile signal over time, test coverage growth from `/ship` auto-generation
 
 ### Code Quality Signals
 - Test LOC ratio trend
diff --git a/ship/SKILL.md b/ship/SKILL.md
index 6ad69ba7..344c35ec 100644
--- a/ship/SKILL.md
+++ b/ship/SKILL.md
@@ -1213,7 +1213,34 @@ EOF
 )"
 ```
 
-**Output the PR URL** — then proceed to Step 8.5.
+**Output the PR URL** — then proceed to Step 8.25.
+
+---
+
+## Step 8.25: Ship Log
+
+Append a structured entry to the ship log for `/retro` velocity tracking. This step is fully automatic and must never fail the workflow.
+
+Collect the data from earlier steps and write a single JSONL entry:
+
+```bash
+source <(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)
+~/.claude/skills/gstack/bin/gstack-ship-log '{"ts":"'"$(date -u +%Y-%m-%dT%H:%M:%SZ)"'","version":"VERSION","branch":"'"$BRANCH"'","repo":"'"$SLUG"'","pr_url":"PR_URL","review_findings":N,"review_auto_fixed":N,"greptile_comments":N,"greptile_fixed":N,"greptile_fps":N,"todos_completed":N,"tests_passed":true,"coverage_before":N,"coverage_after":N}'
+```
+
+Substitute the capitalized placeholders from data gathered in earlier steps:
+- `VERSION`: the new version from Step 4
+- `PR_URL`: the PR URL output from Step 8
+- `review_findings`: total issues found in Step 3.5 (0 if none)
+- `review_auto_fixed`: count of auto-fixed issues in Step 3.5
+- `greptile_comments`: total Greptile comments in Step 3.75 (0 if skipped)
+- `greptile_fixed`: Greptile comments fixed in Step 3.75
+- `greptile_fps`: Greptile false positives in Step 3.75
+- `todos_completed`: items marked complete in Step 5.5 (0 if none)
+- `tests_passed`: `true` (always true here — Step 3 would have stopped if tests failed)
+- `coverage_before`/`coverage_after`: test file counts from Step 3.4 (0 if skipped)
+
+If the `gstack-ship-log` script is not found or the write fails, warn once and continue. Never stop the ship workflow for a log failure.
 
 ---
 
diff --git a/ship/SKILL.md.tmpl b/ship/SKILL.md.tmpl
index a748314d..5dc78dbd 100644
--- a/ship/SKILL.md.tmpl
+++ b/ship/SKILL.md.tmpl
@@ -615,7 +615,34 @@ EOF
 )"
 ```
 
-**Output the PR URL** — then proceed to Step 8.5.
+**Output the PR URL** — then proceed to Step 8.25.
+
+---
+
+## Step 8.25: Ship Log
+
+Append a structured entry to the ship log for `/retro` velocity tracking. This step is fully automatic and must never fail the workflow.
+
+Collect the data from earlier steps and write a single JSONL entry:
+
+```bash
+source <(~/.claude/skills/gstack/bin/gstack-slug 2>/dev/null)
+~/.claude/skills/gstack/bin/gstack-ship-log '{"ts":"'"$(date -u +%Y-%m-%dT%H:%M:%SZ)"'","version":"VERSION","branch":"'"$BRANCH"'","repo":"'"$SLUG"'","pr_url":"PR_URL","review_findings":N,"review_auto_fixed":N,"greptile_comments":N,"greptile_fixed":N,"greptile_fps":N,"todos_completed":N,"tests_passed":true,"coverage_before":N,"coverage_after":N}'
+```
+
+Substitute the capitalized placeholders from data gathered in earlier steps:
+- `VERSION`: the new version from Step 4
+- `PR_URL`: the PR URL output from Step 8
+- `review_findings`: total issues found in Step 3.5 (0 if none)
+- `review_auto_fixed`: count of auto-fixed issues in Step 3.5
+- `greptile_comments`: total Greptile comments in Step 3.75 (0 if skipped)
+- `greptile_fixed`: Greptile comments fixed in Step 3.75
+- `greptile_fps`: Greptile false positives in Step 3.75
+- `todos_completed`: items marked complete in Step 5.5 (0 if none)
+- `tests_passed`: `true` (always true here — Step 3 would have stopped if tests failed)
+- `coverage_before`/`coverage_after`: test file counts from Step 3.4 (0 if skipped)
+
+If the `gstack-ship-log` script is not found or the write fails, warn once and continue. Never stop the ship workflow for a log failure.
 
 ---