prizmkit 1.0.94 → 1.0.96

package/bundled/VERSION.json

@@ -1,5 +1,5 @@
 {
-  "frameworkVersion": "1.0.94",
-  "bundledAt": "2026-03-23T00:08:19.504Z",
-  "bundledFrom": "b1e871c"
+  "frameworkVersion": "1.0.96",
+  "bundledAt": "2026-03-23T12:42:15.546Z",
+  "bundledFrom": "e29472c"
 }

package/bundled/agents/prizm-dev-team-dev.md

@@ -31,8 +31,7 @@ If the snapshot does not exist:
 
 | Path | Purpose |
 |------|---------|
-| `.prizm-docs/` | Architecture index — module structure, interfaces, dependencies, known traps (TRAPS) |
-| `CLAUDE.md` / `CODEBUDDY.md` + `memory/MEMORY.md` | Project memory — development decisions (DECISIONS), interface conventions, project-level rules |
+| `.prizm-docs/` | Architecture index — module structure, interfaces, dependencies, known traps (TRAPS), design decisions (DECISIONS) |
 | `.prizmkit/specs/###-feature-name/` | Feature artifacts — spec.md / plan.md (with Tasks section) |
 
 ### Must Do (MUST)

package/bundled/agents/prizm-dev-team-reviewer.md

@@ -24,8 +24,7 @@ Project documentation is in `.prizm-docs/`. Before review, read `context-snapsho
 
 | Path | Purpose |
 |------|---------|
-| `.prizm-docs/` | Architecture index — module structure, interfaces, dependencies, known traps (TRAPS) |
-| `CLAUDE.md` / `CODEBUDDY.md` + `memory/MEMORY.md` | Project memory — development decisions (DECISIONS), interface conventions, project-level rules |
+| `.prizm-docs/` | Architecture index — module structure, interfaces, dependencies, known traps (TRAPS), design decisions (DECISIONS) |
 | `.prizmkit/specs/###-feature-name/` | Feature artifacts — spec.md / plan.md (with Tasks section) |
 
 ### Must Do (MUST)

package/bundled/dev-pipeline/README.md

@@ -667,37 +667,6 @@ Use `--dry-run` to verify which model will be used without spawning a session:
 
 > **Note**: `--model` support depends on the CLI. `claude` and `cbc` fully support it. If the CLI doesn't support it, the flag is silently ignored.
 
-### Auto-Detection of Available Models
-
-The pipeline automatically detects which AI models are available for your CLI:
-
-```bash
-# Manual detection
-./dev-pipeline/scripts/detect-models.sh
-
-# Results saved to
-cat .prizmkit/available-models.json
-```
-
-Detection methods vary by platform:
-
-- **CodeBuddy (cbc)**: Probes the CLI backend for the full list of supported models
-- **Claude Code**: Self-reports the default model (model switching not available)
-
-The pipeline runs detection automatically on:
-
-- `./run.sh run` — before processing features
-- `./retry-feature.sh` — before retrying
-- `git pull` — via post-merge hook (background, non-blocking)
-
-### Model Validation
-
-When `available-models.json` exists, the pipeline validates feature model fields:
-
-- Warns if a specified model is not in the available list
-- Warns if the CLI doesn't support `--model` switching
-- **Never blocks** — validation is advisory only
-
 ---
 
 ## AI CLI Configuration
@@ -894,7 +863,6 @@ run-bugfix.sh main loop
 ```
 dev-pipeline/bugfix-state/          # Runtime state (gitignored)
 +-- pipeline.json
-+-- current-session.json
 +-- bugs/B-XXX/
     +-- status.json
     +-- sessions/B-XXX-YYYYMMDDHHMMSS/
@@ -971,7 +939,6 @@ dev-pipeline/
 |
 +-- state/                           # Feature pipeline runtime state (gitignored)
 |   +-- pipeline.json
-|   +-- current-session.json
 |   +-- .pipeline.pid                # Daemon PID file
 |   +-- .pipeline-meta.json          # Daemon metadata
 |   +-- pipeline-daemon.log          # Daemon log (50MB rotation)

package/bundled/dev-pipeline/retry-bug.sh

@@ -175,9 +175,6 @@ python3 "$SCRIPTS_DIR/update-bug-status.py" \
     log_warn "Failed to clean bug artifacts (continuing with fresh session only)"
 }
 
-# Auto-detect available models (must run before any git commit operations)
-bash "$SCRIPT_DIR/scripts/detect-models.sh" --quiet 2>/dev/null || true
-
 # ============================================================
 # Generate bootstrap prompt
 # ============================================================

package/bundled/dev-pipeline/retry-feature.sh

@@ -170,9 +170,6 @@ python3 "$SCRIPTS_DIR/update-feature-status.py" \
     log_warn "Failed to clean feature artifacts (continuing with fresh session only)"
 }
 
-# Auto-detect available models (quiet, non-blocking)
-bash "$SCRIPT_DIR/scripts/detect-models.sh" --quiet 2>/dev/null || true
-
 # ============================================================
 # Generate bootstrap prompt
 # ============================================================

package/bundled/dev-pipeline/run-bugfix.sh

@@ -534,9 +534,6 @@ main() {
         log_info "Resuming existing bugfix pipeline..."
     fi
 
-    # Auto-detect available models (must run before any git commit operations)
-    bash "$SCRIPT_DIR/scripts/detect-models.sh" --quiet 2>/dev/null || true
-
     # Print header
     echo ""
     echo -e "${BOLD}════════════════════════════════════════════════════${NC}"
@@ -590,7 +587,6 @@ main() {
             log_success "  All bugs processed! Bug fix pipeline finished."
             log_success "  Total sessions: $session_count"
             log_success "════════════════════════════════════════════════════"
-            rm -f "$STATE_DIR/current-session.json"
 
             # Merge dev branch back to original
             if [[ -n "$_DEV_BRANCH_NAME" ]]; then
@@ -648,23 +644,6 @@ main() {
             --state-dir "$STATE_DIR" \
             --output "$bootstrap_prompt" >/dev/null 2>&1
 
-        # Track current session (atomic write via temp file)
-        python3 -c "
-import json, sys, os
-from datetime import datetime, timezone
-bug_id, session_id, state_dir = sys.argv[1], sys.argv[2], sys.argv[3]
-data = {
-    'bug_id': bug_id,
-    'session_id': session_id,
-    'started_at': datetime.now(timezone.utc).strftime('%Y-%m-%dT%H:%M:%SZ')
-}
-target = os.path.join(state_dir, 'current-session.json')
-tmp = target + '.tmp'
-with open(tmp, 'w') as f:
-    json.dump(data, f, indent=2)
-os.replace(tmp, target)
-" "$bug_id" "$session_id" "$STATE_DIR"
-
         # Spawn session
         log_info "Spawning AI CLI session: $session_id"
         _SPAWN_RESULT=""

package/bundled/dev-pipeline/run.sh

@@ -836,14 +836,6 @@ main() {
         log_info "Resuming existing pipeline..."
     fi
 
-    # Auto-detect available models + validate feature model fields
-    bash "$SCRIPT_DIR/scripts/detect-models.sh" --quiet 2>/dev/null || true
-    if [[ -f ".prizmkit/available-models.json" ]]; then
-        python3 "$SCRIPTS_DIR/validate-feature-models.py" \
-            --feature-list "$feature_list" \
-            --models-file ".prizmkit/available-models.json" 2>&1 | head -5 || true
-    fi
-
     # Print header
     echo ""
     echo -e "${BOLD}════════════════════════════════════════════════════${NC}"
@@ -909,7 +901,6 @@ for f in data.get('stuck_features', []):
             log_success "  All features completed! Pipeline finished."
             log_success "  Total sessions: $session_count"
             log_success "════════════════════════════════════════════════════"
-            rm -f "$STATE_DIR/current-session.json"
             break
         fi
 
@@ -993,23 +984,6 @@ for f in data.get('stuck_features', []):
         local feature_model
         feature_model=$(echo "$gen_output" | python3 -c "import json,sys; print(json.load(sys.stdin).get('model',''))" 2>/dev/null || echo "")
 
-        # Update current session tracking (atomic write via temp file)
-        python3 -c "
-import json, sys, os
-from datetime import datetime, timezone
-feature_id, session_id, state_dir = sys.argv[1], sys.argv[2], sys.argv[3]
-data = {
-    'feature_id': feature_id,
-    'session_id': session_id,
-    'started_at': datetime.now(timezone.utc).strftime('%Y-%m-%dT%H:%M:%SZ')
-}
-target = os.path.join(state_dir, 'current-session.json')
-tmp = target + '.tmp'
-with open(tmp, 'w') as f:
-    json.dump(data, f, indent=2)
-os.replace(tmp, target)
-" "$feature_id" "$session_id" "$STATE_DIR"
-
         # Mark feature as in-progress before spawning session
         python3 "$SCRIPTS_DIR/update-feature-status.py" \
             --feature-list "$feature_list" \
@@ -1036,11 +1010,36 @@ os.replace(tmp, target)
             else
                 log_warn "Auto-merge failed — dev branch preserved: $_DEV_BRANCH_NAME"
                 log_warn "Merge manually: git checkout $_ORIGINAL_BRANCH && git rebase $_DEV_BRANCH_NAME"
+                # branch_merge failed and left us on dev branch — checkout original and commit pipeline artifacts
+                if git -C "$_proj_root" checkout "$_ORIGINAL_BRANCH" 2>/dev/null; then
+                    local _merge_failed_current _merge_failed_dirty
+                    _merge_failed_current=$(git -C "$_proj_root" rev-parse --abbrev-ref HEAD 2>/dev/null || true)
+                    if [[ "$_merge_failed_current" == "$_ORIGINAL_BRANCH" ]]; then
+                        _merge_failed_dirty=$(git -C "$_proj_root" status --porcelain 2>/dev/null || true)
+                        if [[ -n "$_merge_failed_dirty" ]]; then
+                            git -C "$_proj_root" add -A 2>/dev/null || true
+                            git -C "$_proj_root" commit --no-verify -m "chore: include pipeline state artifacts" 2>/dev/null || true
+                        fi
+                    fi
+                fi
                 _DEV_BRANCH_NAME=""
             fi
         elif [[ -n "$_DEV_BRANCH_NAME" ]]; then
             # Session failed — return to original branch, preserve dev branch for inspection
-            git -C "$_proj_root" checkout "$_ORIGINAL_BRANCH" 2>/dev/null || true
+            if git -C "$_proj_root" checkout "$_ORIGINAL_BRANCH" 2>/dev/null; then
+                # Commit any dirty pipeline state files left on the original branch
+                local _failed_current _failed_dirty
+                _failed_current=$(git -C "$_proj_root" rev-parse --abbrev-ref HEAD 2>/dev/null || true)
+                if [[ "$_failed_current" == "$_ORIGINAL_BRANCH" ]]; then
+                    _failed_dirty=$(git -C "$_proj_root" status --porcelain 2>/dev/null || true)
+                    if [[ -n "$_failed_dirty" ]]; then
+                        git -C "$_proj_root" add -A 2>/dev/null || true
+                        git -C "$_proj_root" commit --no-verify -m "chore: include pipeline state artifacts" 2>/dev/null || true
+                    fi
+                fi
+            else
+                log_warn "Failed to checkout $_ORIGINAL_BRANCH after session failure — pipeline state may be uncommitted"
+            fi
             log_warn "Session failed — dev branch preserved for inspection: $_DEV_BRANCH_NAME"
             _DEV_BRANCH_NAME=""
         fi

package/bundled/dev-pipeline/scripts/detect-stuck.py

@@ -188,8 +188,8 @@ def check_stuck_checkpoint(feature_dir):
 def check_stale_heartbeat(feature_id, feature_status, state_dir, stale_threshold):
     """Check 3: Is the heartbeat stale or missing for an in_progress feature?
 
-    Only applies to features whose status is 'in_progress' and whose session
-    matches the current session.
+    Only applies to features whose status is 'in_progress'.
+    Uses last_session_id from the feature's own status to find the active session.
 
     Returns a stuck-report dict or None.
     """
@@ -197,18 +197,8 @@ def check_stale_heartbeat(feature_id, feature_status, state_dir, stale_threshold
     if status != "in_progress":
         return None
 
-    # Read current-session.json to find the active session
-    current_session_path = os.path.join(state_dir, "current-session.json")
-    current_session = load_json(current_session_path)
-    if current_session is None:
-        return None
-
-    # Check if the current session is for this feature
-    session_feature = current_session.get("feature_id")
-    if session_feature != feature_id:
-        return None
-
-    session_id = current_session.get("session_id")
+    # Use last_session_id from the feature's own status
+    session_id = feature_status.get("last_session_id")
     if not session_id:
         return None
 

package/bundled/dev-pipeline/scripts/parse-stream-progress.py

@@ -58,7 +58,7 @@ PHASE_KEYWORDS = {
     },
     "retrospective": {
         "strong": ["prizmkit-retrospective"],
-        "weak": ["retrospective", "structural sync", "knowledge distillation", ".prizm-docs/ sync", "memory sedimentation"],
+        "weak": ["retrospective", "structural sync", "architecture knowledge", ".prizm-docs/ sync"],
     },
     "commit": {
         "strong": ["prizmkit-committer"],

package/bundled/dev-pipeline/scripts/update-bug-status.py

@@ -365,13 +365,7 @@ def cleanup_bug_artifacts(state_dir, bug_id, project_root=None):
         shutil.rmtree(dev_team_dir)
         cleaned.append("Deleted {} ({} files)".format(dev_team_dir, file_count))
 
-    # 5) Clear current-session pointer if it points to this bug
-    current_session_path = os.path.join(state_dir, "current-session.json")
-    if os.path.isfile(current_session_path):
-        current_session, _ = load_json_file(current_session_path)
-        if current_session and current_session.get("bug_id") == bug_id:
-            os.remove(current_session_path)
-            cleaned.append("Deleted {}".format(current_session_path))
+    # 5) (removed: current-session.json no longer used)
 
     return cleaned
 
@@ -603,13 +597,7 @@ def action_clean(args, bug_list_path, state_dir):
         shutil.rmtree(dev_team_dir)
         cleaned.append("Deleted {} ({} files)".format(dev_team_dir, file_count))
 
-    # 4. Delete current-session pointer if it points to this bug
-    current_session_path = os.path.join(state_dir, "current-session.json")
-    if os.path.isfile(current_session_path):
-        current_session, _ = load_json_file(current_session_path)
-        if current_session and current_session.get("bug_id") == bug_id:
-            os.remove(current_session_path)
-            cleaned.append("Deleted {}".format(current_session_path))
+    # 4. (removed: current-session.json no longer used)
 
     # 5. Reset status
     bs = load_bug_status(state_dir, bug_id)

package/bundled/dev-pipeline/scripts/update-feature-status.py

@@ -264,12 +264,7 @@ def cleanup_feature_artifacts(feature_list_path, state_dir, feature_id, project_
         cleaned.append("Deleted {} ({} files)".format(dev_team_dir, file_count))
 
     # 5) Clear current-session pointer if it points to this feature
-    current_session_path = os.path.join(state_dir, "current-session.json")
-    if os.path.isfile(current_session_path):
-        current_session, _ = load_json_file(current_session_path)
-        if current_session and current_session.get("feature_id") == feature_id:
-            os.remove(current_session_path)
-            cleaned.append("Deleted {}".format(current_session_path))
+    # (no-op: current-session.json has been removed from the pipeline)
 
     return cleaned
 
@@ -989,13 +984,7 @@ def action_clean(args, feature_list_path, state_dir):
         shutil.rmtree(dev_team_dir)
         cleaned.append("Deleted {} ({} files)".format(dev_team_dir, file_count))
 
-    # 4. Delete current-session pointer if it points to this feature
-    current_session_path = os.path.join(state_dir, "current-session.json")
-    if os.path.isfile(current_session_path):
-        current_session, _ = load_json_file(current_session_path)
-        if current_session and current_session.get("feature_id") == feature_id:
-            os.remove(current_session_path)
-            cleaned.append("Deleted {}".format(current_session_path))
+    # 4. (removed: current-session.json no longer used)
 
     # 5. Reset status (reuse reset logic)
     fs = load_feature_status(state_dir, feature_id)

package/bundled/dev-pipeline/templates/bootstrap-tier1.md

@@ -133,11 +133,10 @@ Key decisions: [list]
 
 ### Phase 4: Architecture Sync & Commit
 
-**4a.** Run `/prizmkit-retrospective` — maintains `.prizm-docs/` (architecture index) and platform memory files:
+**4a.** Run `/prizmkit-retrospective` — maintains `.prizm-docs/` (architecture index):
 1. **Structural sync**: Use `git diff --cached --name-status` to locate changed modules, update KEY_FILES/INTERFACES/DEPENDENCIES/file counts in affected `.prizm-docs/` files
-2. **Architecture knowledge** (feature sessions only): Extract TRAPS/RULES from completed work into `.prizm-docs/`
-3. **Memory sedimentation** (feature sessions only): Sediment DECISIONS and interface conventions to platform memory file (`CLAUDE.md` for Claude Code, BOTH `CODEBUDDY.md` AND `memory/MEMORY.md` for CodeBuddy)
-4. Stage all doc changes: `git add .prizm-docs/`
+2. **Architecture knowledge** (feature sessions only): Extract TRAPS/RULES/DECISIONS from completed work into `.prizm-docs/`
+3. Stage all doc changes: `git add .prizm-docs/`
 
 Doc maintenance pass condition (pipeline-enforced): `.prizm-docs/` changed in the final commit.
 

package/bundled/dev-pipeline/templates/bootstrap-tier2.md

@@ -180,11 +180,10 @@ If GATE:MISSING — send message to Reviewer (re-spawn if needed): "Write the '#
 
 ### Phase 5: Architecture Sync & Commit
 
-**5a.** Run `/prizmkit-retrospective` — maintains `.prizm-docs/` (architecture index) and platform memory files:
+**5a.** Run `/prizmkit-retrospective` — maintains `.prizm-docs/` (architecture index):
 1. **Structural sync**: Use `git diff --cached --name-status` to locate changed modules, update KEY_FILES/INTERFACES/DEPENDENCIES/file counts in affected `.prizm-docs/` files
-2. **Architecture knowledge** (feature sessions only): Extract TRAPS/RULES from completed work into `.prizm-docs/`
-3. **Memory sedimentation** (feature sessions only): Sediment DECISIONS and interface conventions to platform memory file (`CLAUDE.md` for Claude Code, BOTH `CODEBUDDY.md` AND `memory/MEMORY.md` for CodeBuddy)
-4. Stage all doc changes: `git add .prizm-docs/`
+2. **Architecture knowledge** (feature sessions only): Extract TRAPS/RULES/DECISIONS from completed work into `.prizm-docs/`
+3. Stage all doc changes: `git add .prizm-docs/`
 
 Doc maintenance pass condition (pipeline-enforced): `.prizm-docs/` changed in the final commit.
 

package/bundled/dev-pipeline/templates/bootstrap-tier3.md

@@ -354,12 +354,11 @@ git log --oneline | grep "{{FEATURE_ID}}" | head -3
 - If a commit for `{{FEATURE_ID}}` already exists → **skip 6c** (do NOT run /prizmkit-committer, do NOT run git reset, do NOT stage or unstage anything). Proceed directly to Final Clean Check.
 - If no existing commit → proceed normally with 6a–6c.
 
-**6b.** Run `/prizmkit-retrospective` (**before commit**, maintains `.prizm-docs/` architecture index and platform memory files):
+**6b.** Run `/prizmkit-retrospective` (**before commit**, maintains `.prizm-docs/` architecture index):
 - **Structural sync**: update KEY_FILES/INTERFACES/DEPENDENCIES/file counts for changed modules
-- **Architecture knowledge** (feature sessions only): extract TRAPS, RULES from completed work into `.prizm-docs/`
-- **Memory sedimentation** (feature sessions only): sediment DECISIONS and interface conventions to platform memory file (`CLAUDE.md` for Claude Code, BOTH `CODEBUDDY.md` AND `memory/MEMORY.md` for CodeBuddy)
+- **Architecture knowledge** (feature sessions only): extract TRAPS, RULES, DECISIONS from completed work into `.prizm-docs/`
 - Stage all doc changes: `git add .prizm-docs/`
-- **For bug-fix sessions**: structural sync only, skip knowledge injection and memory sedimentation unless a genuinely new pitfall was discovered
+- **For bug-fix sessions**: structural sync only, skip knowledge injection unless a genuinely new pitfall was discovered
 
 **6c.** Run `/prizmkit-committer` → `feat({{FEATURE_ID}}): {{FEATURE_TITLE}}`, do NOT push
 

package/bundled/dev-pipeline/templates/feature-list-schema.json

@@ -99,7 +99,7 @@
           },
           "model": {
             "type": "string",
-            "description": "AI model ID for this feature. Overrides $MODEL env var. See .prizmkit/available-models.json for valid IDs."
+            "description": "AI model ID for this feature. Overrides $MODEL env var."
           }
         }
       }

package/bundled/rules/prizm/prizm-commit-workflow.md

@@ -3,7 +3,7 @@ description: "PrizmKit commit workflow rules"
 ---
 
 Before any git commit in this project:
-1. Run `/prizmkit-retrospective` to sync `.prizm-docs/` (architecture index) and sediment DECISIONS to memory files
+1. Run `/prizmkit-retrospective` to sync `.prizm-docs/` (architecture index with TRAPS/RULES/DECISIONS)
 2. Use Conventional Commits format: type(scope): description
 3. Bug fixes use `fix()` prefix, not `feat()`
 4. Bug fixes run retrospective with structural sync only (Job 1)

package/bundled/skills/_metadata.json

@@ -1,5 +1,5 @@
 {
-  "version": "1.0.94",
+  "version": "1.0.96",
   "skills": {
     "prizm-kit": {
       "description": "Full-lifecycle dev toolkit. Covers spec-driven development, Prizm context docs, code quality, debugging, deployment, and knowledge management.",
@@ -65,7 +65,7 @@
       "hasScripts": false
     },
     "prizmkit-retrospective": {
-      "description": "Sole .prizm-docs/ maintainer. Structural sync + TRAPS/RULES injection to architecture index. Sediments DECISIONS to platform memory files. Run after code review passes, before committing.",
+      "description": "Sole .prizm-docs/ maintainer. Structural sync + TRAPS/RULES/DECISIONS injection to architecture index. Run after code review passes, before committing.",
       "tier": "1",
       "category": "prizmkit-skill",
       "hasAssets": true,
@@ -255,7 +255,7 @@
       ]
     },
     "minimal": {
-      "description": "Minimal closed-loop workflow: full spec-driven cycle with project memory and code review",
+      "description": "Minimal closed-loop workflow: full spec-driven cycle with .prizm-docs/ architecture sync and code review",
       "skills": [
         "prizm-kit",
         "prizmkit-init",

package/bundled/skills/app-planner/SKILL.md

@@ -49,8 +49,6 @@ Before questions, check optional context files (never block if absent):
 - `.prizm-docs/root.prizm` (architecture/project context)
 - `.prizmkit/config.json` (existing stack preferences)
 - existing `feature-list.json` (required for incremental mode)
-- `.prizmkit/available-models.json` (detected AI models, from detect-models.sh)
-
 Note:
 - This skill **reads** `.prizmkit/config.json` if present.
 - This skill does **not** create `.prizmkit/config.json` directly.
@@ -84,12 +82,6 @@ Execute the selected scenario workflow in conversation mode with mandatory check
 ### Interactive Phases
 1. clarify business goal and scope
 2. confirm constraints and tech assumptions
-2b. (optional) if `.prizmkit/available-models.json` exists and `model_switch_supported == true`:
-    - read the file and show available models to user
-    - ask: "要为 feature 指定 AI 模型吗？" with options:
-      a) 全部使用某个模型（写入每个 feature 的 model 字段）
-      b) 不指定（使用 $MODEL 环境变量或 CLI 默认）
-    - if user chooses a model, remember it for Phase 6 JSON generation
 3. propose feature set with dependencies
 4. refine descriptions and acceptance criteria
 5. verify DAG/order/priorities
@@ -158,11 +150,6 @@ AI: "Ready to proceed to dev-pipeline."
 - valid dependency DAG
 - new items default `status: "pending"`
 - English feature titles for stable slug generation
-- if `.prizmkit/available-models.json` exists and `model_switch_supported == true`:
-  - if user specified model preference during planning:
-    - set `model` field on every feature with user's chosen model
-  - if user said "no preference" → omit `model` field
-- model IDs should come from `.prizmkit/available-models.json` models list
 - `model` field is optional — omitting it means the pipeline uses $MODEL env or CLI default
 
 ## Next-Step Execution Policy (after planning)

package/bundled/skills/bugfix-pipeline-launcher/SKILL.md

@@ -189,8 +189,8 @@ Detect user intent from their message, then follow the corresponding workflow:
 
 4. **For per-bug session logs** (when user asks about a specific bug):
    ```bash
-   # Find current/recent session
-   cat dev-pipeline/bugfix-state/current-session.json 2>/dev/null
+   # Check bug status for last session ID
+   cat dev-pipeline/bugfix-state/bugs/<BUG_ID>/status.json 2>/dev/null
    # Then tail that bug's session log
    tail -100 dev-pipeline/bugfix-state/bugs/<BUG_ID>/sessions/<SESSION_ID>/logs/session.log
    ```

package/bundled/skills/dev-pipeline-launcher/SKILL.md

@@ -225,8 +225,8 @@ Detect user intent from their message, then follow the corresponding workflow:
 
 4. **For per-feature session logs** (when user asks about a specific feature):
    ```bash
-   # Find current/recent session
-   cat dev-pipeline/state/current-session.json 2>/dev/null
+   # Check feature status for last session ID
+   cat dev-pipeline/state/features/<FEATURE_ID>/status.json 2>/dev/null
    # Then tail that feature's session log
    tail -100 dev-pipeline/state/features/<FEATURE_ID>/sessions/<SESSION_ID>/logs/session.log
    ```
@@ -296,4 +296,4 @@ Notes:
 - **Single instance**: Only one pipeline can run at a time. The PID file prevents duplicates.
 - **Pipeline coexistence**: Feature and bugfix pipelines use separate state directories (`state/` vs `bugfix-state/`), so they can run simultaneously without conflict.
 - **State preservation**: Stopping and restarting the pipeline resumes from where it left off -- completed features are not re-run.
-- **HANDOFF**: After pipeline completes all features, suggest running `prizmkit-retrospective` for project memory update, or ask user what's next.
+- **HANDOFF**: After pipeline completes all features, suggest running `prizmkit-retrospective` for `.prizm-docs/` architecture sync, or ask user what's next.

package/bundled/skills/prizm-kit/SKILL.md

@@ -74,17 +74,15 @@ All documentation records are for: features, projects, code logic, and module in
 
 ## Architecture
 
-PrizmKit produces three complementary knowledge layers:
+PrizmKit produces two complementary knowledge layers:
 
 ```
-.prizm-docs/           → Architecture Index (structure, interfaces, dependencies, traps, rules)
+.prizm-docs/           → Architecture Index (structure, interfaces, dependencies, traps, rules, decisions)
 .prizmkit/specs/       → Feature "what to do" (workflow: spec → plan → code(implement))
-CLAUDE.md / CODEBUDDY.md → Project Memory (decisions, interface conventions, project-level rules)
 ```
 
 **Reading guide**:
-- Need code structure/modules/interfaces/traps? → `.prizm-docs/`
-- Need past decisions/conventions/why? → `CLAUDE.md` (Claude Code) or `CODEBUDDY.md` + `memory/MEMORY.md` (CodeBuddy)
+- Need code structure/modules/interfaces/traps/decisions? → `.prizm-docs/`
 
 ## Skill Inventory
 
@@ -100,7 +98,7 @@ CLAUDE.md / CODEBUDDY.md → Project Memory (decisions, interface conventions, p
 - **prizmkit-analyze** — Cross-document consistency analysis (spec ↔ plan ↔ tasks)
 - **prizmkit-implement** — Execute tasks following TDD approach
 - **prizmkit-code-review** — Review code against spec and plan
-- **prizmkit-retrospective** — Sole .prizm-docs/ maintainer: structural sync + TRAPS/RULES injection. Also sediments DECISIONS to platform memory files
+- **prizmkit-retrospective** — Sole .prizm-docs/ maintainer: structural sync + TRAPS/RULES/DECISIONS injection
 - **prizmkit-committer** — Pure git commit: diff analysis, safety checks, Conventional Commits
 
 ### Quality Assurance (5)

package/bundled/skills/prizm-kit/assets/project-memory-template.md

@@ -5,7 +5,7 @@ This project uses PrizmKit with the Prizm documentation system for AI-optimized
 ### Progressive Loading Protocol
 - ON SESSION START: Always read `.prizm-docs/root.prizm` first (L0 — project map)
 - ON TASK: Read L1 (`.prizm-docs/<module>.prizm`) for relevant modules referenced in MODULE_INDEX
-- ON FILE EDIT: Read L2 (`.prizm-docs/<module>/<submodule>.prizm`) before modifying files. Pay attention to TRAPS.
+- ON FILE EDIT: Read L2 (`.prizm-docs/<module>/<submodule>.prizm`) before modifying files. Pay attention to TRAPS and DECISIONS.
 - NEVER load all .prizm docs at once. Load only what is needed for the current task.
 
 ### Auto-Update Protocol

package/bundled/skills/prizmkit-code-review/SKILL.md

@@ -26,7 +26,7 @@ Perform a comprehensive code review against the feature spec, implementation pla
    - If `fix-plan.md` exists → **Bugfix mode**: read `fix-plan.md` as baseline. Review dimensions: bug is actually fixed, no regressions, reproduction test passes, minimal change scope.
    - If none found → prompt user: "No review baseline found. Which workflow are you in? (feature/refactor/bugfix)"
 2. Read **architecture index**: `.prizm-docs/root.prizm` RULES and PATTERNS for project conventions
-3. Read **project memory**: platform memory file (`CLAUDE.md` for Claude Code, `CODEBUDDY.md` + `memory/MEMORY.md` for CodeBuddy) for past DECISIONS — helps verify implementation respects established conventions
+3. Read **past decisions**: check DECISIONS sections in relevant `.prizm-docs/` L1/L2 files — helps verify implementation respects established conventions
 4. Read '## Implementation Log' section of context-snapshot.md in the feature directory (if exists) — understand Dev's implementation decisions, trade-offs, and notable discoveries. This context helps distinguish intentional design choices from accidental patterns during review.
 5. Scan all code files referenced in completed tasks
 4. Review across 6 dimensions:

package/bundled/skills/prizmkit-committer/SKILL.md

@@ -1,17 +1,17 @@
 ---
 name: "prizmkit-committer"
-description: "Pure git commit workflow with safety checks. Stages files, analyzes diff, generates Conventional Commits message, and commits. Does NOT modify .prizm-docs/ or memory files — architecture sync and memory sedimentation are handled by /prizmkit-retrospective before this skill is invoked. Trigger on: 'commit', 'submit', 'finish', 'done', 'ship it', 'save my work'. (project)"
+description: "Pure git commit workflow with safety checks. Stages files, analyzes diff, generates Conventional Commits message, and commits. Does NOT modify .prizm-docs/ — architecture sync is handled by /prizmkit-retrospective before this skill is invoked. Trigger on: 'commit', 'submit', 'finish', 'done', 'ship it', 'save my work'. (project)"
 ---
 
 # PrizmKit Committer
 
 Pure git commit workflow. Analyzes changes, generates a Conventional Commits message, performs safety checks, and commits.
 
-**This skill is a pure git commit tool. It does NOT modify any project files — no `.prizm-docs/`, no memory files, no source code.** It only reads diffs, generates a commit message, and commits. For feature/refactor workflows, run `/prizmkit-retrospective` before this skill to sync `.prizm-docs/` (architecture index) and sediment DECISIONS to memory files. For bug fixes, skip retrospective entirely — bug fixes do not update `.prizm-docs/`.
+**This skill is a pure git commit tool. It does NOT modify any project files — no `.prizm-docs/`, no source code.** It only reads diffs, generates a commit message, and commits. For feature/refactor workflows, run `/prizmkit-retrospective` before this skill to sync `.prizm-docs/` (architecture index with TRAPS/RULES/DECISIONS). For bug fixes, skip retrospective entirely — bug fixes do not update `.prizm-docs/`.
 
 ### When to Use
 - User says "commit", "submit", "finish", "done with this task", "ship it"
-- After `/prizmkit-retrospective` has finished architecture sync and memory sedimentation
+- After `/prizmkit-retrospective` has finished architecture sync
 - The UserPromptSubmit hook will remind to use this skill when commit intent is detected
 
 ### Workflow

package/bundled/skills/prizmkit-implement/SKILL.md

@@ -30,9 +30,8 @@ Execute implementation by following the task breakdown in plan.md. Respects task
 2. Load project context — use the most efficient source available:
    - If `context-snapshot.md` exists in the feature directory → read it. Section 3 has Prizm docs + TRAPS. Section 4 has File Manifest (Tier-2/3) or full source (Tier-1). Read source files on-demand as directed by the manifest.
    - Otherwise → **self-service context fallback**:
-     1. Read **architecture index**: `.prizm-docs/root.prizm` and relevant L1/L2 for affected modules. Pay special attention to TRAPS.
-     2. Read **project memory**: platform memory file for past DECISIONS and interface conventions.
-     3. Scan needed source files
+     1. Read **architecture index**: `.prizm-docs/root.prizm` and relevant L1/L2 for affected modules. Pay special attention to TRAPS and DECISIONS.
+     2. Scan needed source files
 3. Check if checkpoint tasks are complete before proceeding to next phase
 4. For each unchecked task in order:
    a. If task has `[P]` marker, it can run in parallel with other `[P]` tasks in the same group