prizmkit 1.0.35 → 1.0.58

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

@@ -13,7 +13,7 @@ You are the **session orchestrator**. Implement Feature {{FEATURE_ID}}: "{{FEATU
 
 **CRITICAL**: You MUST NOT exit until ALL work is complete and session-status.json is written. When you spawn subagents, wait for each to finish (run_in_background=false). Do NOT spawn agents in background and exit — that kills the session.
 
-**Tier 3 — Full Team**: PM + Dev + Reviewer agents spawned directly via Task tool. Full 7-phase pipeline.
+**Tier 3 — Full Team**: For complex features, create a coordinated agent team via TeamCreate. Use the full 7-phase pipeline with Dev + Reviewer agents.
 
 ### Feature Description
 
@@ -68,16 +68,14 @@ LLM context is frozen at prompt time. Modifying a skill source file during this
 ## PrizmKit Directory Convention
 
 ```
-.prizmkit/specs/{{FEATURE_SLUG}}/context-snapshot.md  ← PM writes, all agents read
+.prizmkit/specs/{{FEATURE_SLUG}}/context-snapshot.md  ← orchestrator writes, all subagents read
 .prizmkit/specs/{{FEATURE_SLUG}}/spec.md
 .prizmkit/specs/{{FEATURE_SLUG}}/plan.md              ← includes Tasks section
-.prizmkit/specs/REGISTRY.md
 ```
 
-**`context-snapshot.md`** is the shared knowledge base. PM writes it once; Dev and Reviewer read it instead of re-scanning source files. This eliminates redundant I/O across all agents.
+**`context-snapshot.md`** is the shared knowledge base. Orchestrator writes it once; Dev and Reviewer read it instead of re-scanning source files. This eliminates redundant I/O across all agents.
 
 ### Agent Files
-- PM Agent: `{{PM_SUBAGENT_PATH}}`
 - Dev Agent: `{{DEV_SUBAGENT_PATH}}`
 - Reviewer Agent: `{{REVIEWER_SUBAGENT_PATH}}`
 
@@ -120,7 +118,7 @@ Save the list of **pre-existing failing tests** (if any) as `BASELINE_FAILURES`.
 
 ### Step 1: Team Setup
 
-No TeamCreate required. Agents are spawned directly via the `Task` tool using `subagent_type`.
+Create a coordinated agent team for this feature session:
 
 1. Run init script:
    `python3 {{INIT_SCRIPT_PATH}} --project-root {{PROJECT_ROOT}} --feature-id {{FEATURE_ID}} --feature-slug {{FEATURE_SLUG}}`
@@ -128,8 +126,12 @@ No TeamCreate required. Agents are spawned directly via the `Task` tool using `s
 2. Check for existing artifacts:
    `ls .prizmkit/specs/{{FEATURE_SLUG}}/ 2>/dev/null`
 
+3. Create the agent team:
+   - Call `TeamCreate` with `team_name="prizm-dev-team-{{FEATURE_ID}}"` and `description="Implementing {{FEATURE_TITLE}}"`
+   - Reference team config at `{{TEAM_CONFIG_PATH}}` for agent definitions
+   - Spawn Dev and Reviewer agents as team members
+
 Agent files are at:
-- PM: `{{PM_SUBAGENT_PATH}}`
 - Dev: `{{DEV_SUBAGENT_PATH}}`
 - Reviewer: `{{REVIEWER_SUBAGENT_PATH}}`
 
@@ -144,7 +146,7 @@ python3 {{INIT_SCRIPT_PATH}} --project-root {{PROJECT_ROOT}} --feature-id {{FEAT
 After team setup: check `.prizmkit/specs/{{FEATURE_SLUG}}/context-snapshot.md` — if exists, all agents MUST use it. Read existing artifacts and resume from Phase {{RESUME_PHASE}}.
 {{END_IF_RESUME}}
 
-### Phase 1-2: Specify + Plan — PM Agent
+### Phase 1-2: Specify + Plan — Orchestrator (you)
 
 Check existing artifacts first:
 ```bash
@@ -152,40 +154,44 @@ ls .prizmkit/specs/{{FEATURE_SLUG}}/ 2>/dev/null
 ```
 
 - Both (spec.md, plan.md) exist → **SKIP to CP-1**
-- `context-snapshot.md` exists → PM reads it instead of re-scanning source files
-- Some missing → PM generates only missing files
+- `context-snapshot.md` exists → use it directly, skip Phase 1
+- Some missing → generate only missing files
 
-Before spawning PM, check whether feature code already exists in the project:
+Before planning, check whether feature code already exists in the project:
 ```bash
 grep -r "{{FEATURE_SLUG}}" src/ --include="*.js" --include="*.ts" -l 2>/dev/null | head -20
 ```
 
-Record result as `EXISTING_CODE` (list of files, or empty). Pass this to PM prompt below.
+Record result as `EXISTING_CODE` (list of files, or empty).
+
+If `EXISTING_CODE` is non-empty: your spec/plan/tasks must reflect this existing implementation — document what exists, identify gaps, do NOT re-implement what is already done.
+
+**Step A — Build Context Snapshot** (skip if `context-snapshot.md` already exists):
 
-Spawn PM agent (Task tool, subagent_type="prizm-dev-team-pm", run_in_background=false).
+1. Read `.prizm-docs/root.prizm` and relevant L1/L2 prizm docs
+2. Scan `src/` for files related to this feature; read each one
+3. Write `.prizmkit/specs/{{FEATURE_SLUG}}/context-snapshot.md`:
+   - **Section 1 — Feature Brief**: feature description + acceptance criteria (copy from above)
+   - **Section 2 — Project Structure**: relevant `ls src/` output
+   - **Section 3 — Prizm Context**: full content of root.prizm and relevant L1/L2 docs
+   - **Section 4 — Existing Source Files**: **full verbatim content** of each related file in fenced code blocks (with `### path/to/file` heading and line count). Include ALL files needed for implementation and review — downstream subagents read this section instead of re-reading individual source files
+   - **Section 5 — Existing Tests**: full content of related test files as code blocks
+4. Confirm: `ls .prizmkit/specs/{{FEATURE_SLUG}}/context-snapshot.md`
 
-**Construct prompt dynamically** — always prefix with:
-> "Read {{PM_SUBAGENT_PATH}}. For feature {{FEATURE_ID}} (slug: {{FEATURE_SLUG}}), complete the following IN THIS SINGLE SESSION — do NOT exit until ALL listed steps are done and files are written to disk:"
+**After Step A, do NOT re-read any original source files** — use context-snapshot.md for all subsequent work.
 
-If `EXISTING_CODE` is non-empty, append to prefix:
-> "NOTE: The following files related to this feature already exist in the codebase: `<EXISTING_CODE list>`. Your spec/plan/tasks must reflect this existing implementation — document what exists, identify gaps, do NOT re-implement what is already done."
+**Step B — Planning Artifacts** (generate only missing files):
 
-**Step A — Build Context Snapshot** (include only if `context-snapshot.md` does NOT exist):
-> "Step A: Write `.prizmkit/specs/{{FEATURE_SLUG}}/context-snapshot.md`. This is the team knowledge base — complete it before anything else. Include:
-> - Section 1 'Feature Brief': feature description and acceptance criteria
-> - Section 2 'Project Structure': output of `ls src/` and relevant subdirectories
-> - Section 3 'Prizm Context': full content of `.prizm-docs/root.prizm` and relevant L1/L2 docs
-> - Section 4 'Existing Source Files': full content of every related source file as a code block
-> - Section 5 'Existing Tests': full content of related test files as code blocks
-> Confirm with `ls .prizmkit/specs/{{FEATURE_SLUG}}/context-snapshot.md`."
+```bash
+ls .prizmkit/specs/{{FEATURE_SLUG}}/spec.md .prizmkit/specs/{{FEATURE_SLUG}}/plan.md 2>/dev/null
+```
 
-**Step B — Planning Artifacts** (include only missing files):
-- spec.md missing: "Run prizmkit-specify → generate spec.md. Resolve any `[NEEDS CLARIFICATION]` markers using the feature description — do NOT pause for interactive input."
-- plan.md missing: "Run prizmkit-plan → generate plan.md (architecture, components, interface design, data model, testing strategy, risk assessment, and Tasks section with `[ ]` checkboxes)"
+- spec.md missing: Run `/prizmkit-specify` → generate spec.md. Resolve any `[NEEDS CLARIFICATION]` markers using the feature description — do NOT pause for interactive input.
+- plan.md missing: Run `/prizmkit-plan` → generate plan.md (architecture, components, interface design, data model, testing strategy, risk assessment, and Tasks section with `[ ]` checkboxes)
 
-> "All files go under `.prizmkit/specs/{{FEATURE_SLUG}}/`. Confirm each with `ls` after writing."
+> All files go under `.prizmkit/specs/{{FEATURE_SLUG}}/`. Confirm each with `ls` after writing.
 
-Wait for PM to return. **CP-1**: Both files exist. If missing, diagnose from PM output — do NOT spawn another PM blindly.
+**CP-1**: Both spec.md and plan.md exist.
 
 ### Phase 4: Analyze — Reviewer Agent
 
@@ -199,9 +205,7 @@ Prompt:
 > Report: CRITICAL, HIGH, MEDIUM issues found (or 'No issues found')."
 
 Wait for Reviewer to return.
-- If CRITICAL issues found: spawn PM to fix — use this prompt:
-  > "Read {{PM_SUBAGENT_PATH}}. Read `.prizmkit/specs/{{FEATURE_SLUG}}/context-snapshot.md` FIRST for full project context. Do NOT re-read individual source files. Fix ONLY the following CRITICAL issues in spec.md/plan.md/tasks.md: `<list issues>`. Do NOT exit until all files are updated."
-  Then re-run analyze (max 1 round).
+- If CRITICAL issues found: fix them yourself — read `.prizmkit/specs/{{FEATURE_SLUG}}/context-snapshot.md` for full project context. Fix ONLY the listed CRITICAL issues in spec.md/plan.md/tasks.md. Then re-run analyze (max 1 round).
 
 **CP-2**: No CRITICAL issues.
 
@@ -287,7 +291,7 @@ Wait for Reviewer to return.
 
 **CP-3**: Integration tests pass, verdict is not NEEDS_FIXES.
 
-### Phase 7: Summarize & Commit — DO NOT SKIP
+### Phase 7: Retrospective & Commit — DO NOT SKIP
 
 {{IF_MODE_SELF_EVOLVE}}
 **Framework Validation Gate (self-evolve mode)**:
@@ -310,7 +314,7 @@ bash {{VALIDATOR_SCRIPTS_DIR}}/validate-framework.sh
 Then set `"reload_needed": true` in session-status.json.
 {{END_IF_MODE_SELF_EVOLVE}}
 
-**For bug fixes**: skip `/prizmkit-summarize`, skip retrospective, and use `fix(<scope>):` commit prefix.
+**For bug fixes**: run `/prizmkit-retrospective` for structural sync only (skip knowledge injection unless a new TRAPS was discovered). Use `fix(<scope>):` commit prefix.
 
 **7a.** Check if feature already committed:
 ```bash
@@ -319,33 +323,17 @@ git log --oneline | grep "{{FEATURE_ID}}" | head -3
 - If a commit for `{{FEATURE_ID}}` already exists → **skip 7c** (do NOT run /prizmkit-committer, do NOT run git reset, do NOT stage or unstage anything). Proceed directly to Step 3.
 - If no existing commit → proceed normally with 7a–7c.
 
-**7b.** Run `/prizmkit-summarize` → archive to REGISTRY.md
+**7b.** Run `/prizmkit-retrospective` (**before commit**, the sole maintainer of `.prizm-docs/`):
+- **Structural sync**: update KEY_FILES/INTERFACES/DEPENDENCIES/file counts for changed modules
+- **Knowledge injection** (feature sessions only): extract TRAPS, RULES, DECISIONS from completed work
+- Stage all doc changes: `git add .prizm-docs/`
+- **For bug-fix sessions**: structural sync only, skip knowledge injection unless a genuinely new pitfall was discovered
 
-**7b.5.** Update prizm-docs (feature sessions):
-- Read the 'Implementation Log' in `context-snapshot.md` to identify changed files
-- For each changed file, check if any related `.prizm-docs/` L1/L2 doc needs updating (new APIs, changed behavior, new TRAPS)
-- Update `.prizm-docs/changelog.prizm`: append one-line entry `- feat({{FEATURE_ID}}): {{FEATURE_TITLE}}`
-- Stage all modified prizm-docs: `git add .prizm-docs/`
-- Pipeline docs pass condition is `REGISTRY.md` OR `.prizm-docs/` changed in final commit
-
-**7b.6.** Run `/prizmkit-retrospective` (feature sessions only, **before commit**):
-- Extract lessons from completed feature — compare spec/plan/tasks vs actual
-- Update relevant `.prizm-docs/` sections: TRAPS, RULES, DECISIONS
-- Write `.prizmkit/specs/{{FEATURE_SLUG}}/retrospective.md`
-- Stage any `.prizm-docs/` changes: `git add .prizm-docs/`
-- **Skip if this session is bug-fix-only**
-
-**7c.** Mark feature complete:
-```bash
-python3 {{VALIDATOR_SCRIPTS_DIR}}/update-feature-status.py \
-  --feature-list "{{FEATURE_LIST_PATH}}" \
-  --state-dir "{{PROJECT_ROOT}}/dev-pipeline/state" \
-  --feature-id "{{FEATURE_ID}}" --session-id "{{SESSION_ID}}" --action complete
-```
+**7c.** Run `/prizmkit-committer` → `feat({{FEATURE_ID}}): {{FEATURE_TITLE}}`, do NOT push
 
-**7d.** Run `/prizmkit-committer` → `feat({{FEATURE_ID}}): {{FEATURE_TITLE}}`, do NOT push
+**7d.** MANDATORY: commit must be done via `/prizmkit-committer` skill. Do NOT run manual `git add`/`git commit` as a substitute.
 
-**7e.** MANDATORY: commit must be done via `/prizmkit-committer` skill. Do NOT run manual `git add`/`git commit` as a substitute.
+**7e.** Do NOT run `update-feature-status.py` here — the pipeline runner handles feature-list.json updates automatically after session exit.
 
 ---
 
@@ -389,18 +377,20 @@ After writing `session-status.json`, verify repository is clean:
 git status --short
 ```
 
-If any files remain, include them in the last commit:
+If any files remain (e.g. session-status.json), stage and create a follow-up commit:
 
 ```bash
 git add -A
-git commit --amend --no-edit
+git commit -m "chore({{FEATURE_ID}}): include session artifacts"
 ```
 
 Re-check `git status --short` and ensure it is empty before exiting.
 
 ## Step 4: Team Cleanup
 
-No team cleanup needed — agents were spawned directly without TeamCreate.
+Clean up the agent team created in Step 1:
+- Call `TeamDelete` to remove the team and free resources
+- Only call TeamDelete after all agents have completed and session-status.json is written
 
 ## Critical Paths
 
@@ -409,7 +399,6 @@ No team cleanup needed — agents were spawned directly without TeamCreate.
 | Feature Artifacts Dir | `.prizmkit/specs/{{FEATURE_SLUG}}/` |
 | Context Snapshot | `.prizmkit/specs/{{FEATURE_SLUG}}/context-snapshot.md` |
 | Team Config | `{{TEAM_CONFIG_PATH}}` |
-| PM Agent Def | {{PM_SUBAGENT_PATH}} |
 | Dev Agent Def | {{DEV_SUBAGENT_PATH}} |
 | Reviewer Agent Def | {{REVIEWER_SUBAGENT_PATH}} |
 | Session Status Output | {{SESSION_STATUS_PATH}} |
@@ -418,8 +407,8 @@ No team cleanup needed — agents were spawned directly without TeamCreate.
 
 ## Reminders
 
-- Tier 3: full team — PM (planning) → Dev (implementation) → Reviewer (review) — agents spawned directly via Task tool (no TeamCreate needed)
-- context-snapshot.md is the team knowledge base: PM writes it once, all agents read it
+- Tier 3: full team — Dev (implementation) → Reviewer (review) — create coordinated team via TeamCreate, clean up with TeamDelete at session end
+- context-snapshot.md is the team knowledge base: orchestrator writes it once, all agents read it
 - Do NOT use `run_in_background=true` when spawning agents
 - ALWAYS write session-status.json before exiting
 - Commit phase must use `/prizmkit-committer`; do NOT replace with manual git commit commands

package/bundled/dev-pipeline/templates/bugfix-bootstrap-prompt.md

@@ -18,17 +18,17 @@ You are the **bug fix session orchestrator**. Fix Bug {{BUG_ID}}: "{{BUG_TITLE}}
 
 **CRITICAL SESSION LIFECYCLE RULE**: You MUST NOT exit until ALL work is complete and session-status.json is written. When you spawn subagents, you MUST **wait for each to finish** (run_in_background=false) before proceeding. Do NOT spawn an agent in the background and exit — that kills the session.
 
-**MANDATORY TEAM REQUIREMENT**: You MUST use the `prizm-dev-team` multi-agent team. This is NON-NEGOTIABLE. All implementation and review work MUST be performed by the appropriate team agents (Dev, Reviewer).
+**MANDATORY TEAM REQUIREMENT**: You MUST use the `prizm-dev-team` multi-agent team. This is NON-NEGOTIABLE. All implementation and review work MUST be performed by the appropriate team agents (Dev, Reviewer). You are the orchestrator — handle coordination, planning, and commit phases directly.
 
 **BUG FIX DOCUMENTATION POLICY**: Bug fixes MUST NOT be recorded as new documentation entries:
-- Do NOT run `/prizmkit-summarize` (no REGISTRY.md entries)
+- Run `/prizmkit-retrospective` with structural sync only (Job 1) — skip knowledge injection unless a genuinely new TRAP was discovered
 - Do NOT create spec/plan/tasks under `.prizmkit/specs/`
 - Do NOT update `.prizm-docs/` module docs for pure bug fixes (unless TRAPS update is needed)
 - Commit with `fix(<scope>):` prefix, NOT `feat:`
 
 ### Team Definition Reference
 
-- **Source of truth**: `core/team/prizm-dev-team.json`
+- **Source of truth**: `{{TEAM_CONFIG_PATH}}`
 - **Installed team config**: `{{TEAM_CONFIG_PATH}}`
 
 ### Bug Description
@@ -81,7 +81,7 @@ You are the **bug fix session orchestrator**. Fix Bug {{BUG_ID}}: "{{BUG_TITLE}}
    - If valid, reuse it. Set `TEAM_REUSED=true`
 
 2. **If no reusable team**, create a new one:
-   - Reference `core/team/prizm-dev-team.json`
+   - Reference `{{TEAM_CONFIG_PATH}}`
    - Call `TeamCreate` with `team_name="prizm-dev-team-{{BUG_ID}}"` and `description="Fixing {{BUG_TITLE}}"`
    - Set `TEAM_REUSED=false`
 
@@ -211,7 +211,7 @@ You are the **bug fix session orchestrator**. Fix Bug {{BUG_ID}}: "{{BUG_TITLE}}
   1. Run `/prizmkit-committer` with:
      - Commit message: `fix({{FIX_SCOPE}}): {{BUG_TITLE}}`
      - Include both fix code and reproduction test
-     - Do NOT run `/prizmkit-summarize`
+     - Do NOT run `/prizmkit-retrospective` with REGISTRY archiving
      - Do NOT push (user will push manually)
   2. If a new pitfall was discovered (not previously in TRAPS):
      - Update the affected module's TRAPS section in `.prizm-docs/`
@@ -267,7 +267,7 @@ Write to: `{{SESSION_STATUS_PATH}}`
 
 | Resource | Path |
 |----------|------|
-| Team Definition (source of truth) | `core/team/prizm-dev-team.json` |
+| Team Definition (source of truth) | `{{TEAM_CONFIG_PATH}}` |
 | Team Config (installed) | `{{TEAM_CONFIG_PATH}}` |
 | Bug Fix Artifacts Dir | `.prizmkit/bugfix/{{BUG_ID}}/` |
 | Fix Plan | `.prizmkit/bugfix/{{BUG_ID}}/fix-plan.md` |
@@ -282,7 +282,7 @@ Write to: `{{SESSION_STATUS_PATH}}`
 - **MANDATORY**: Use `prizm-dev-team` — single-agent execution is FORBIDDEN
 - **Only 2 artifact files per bug**: fix-plan.md + fix-report.md — NEVER more
 - **Do NOT create** spec.md, plan.md, or tasks.md for bug fixes
-- **Do NOT run** `/prizmkit-summarize` (no REGISTRY.md entries for bugs)
+- **Do NOT run** `/prizmkit-retrospective` knowledge injection for bugs (structural sync only, unless genuinely new TRAP discovered)
 - **Commit with** `fix(<scope>):` prefix, NOT `feat:`
 - **Update TRAPS** in `.prizm-docs/` only if a genuinely new pitfall was discovered
 - Dev agents use TDD approach: reproduction test goes from RED → GREEN

package/bundled/dev-pipeline/templates/session-status-schema.json

@@ -15,7 +15,7 @@
     },
     "status": {
       "type": "string",
-      "enum": ["success", "partial", "failed", "commit_missing", "docs_missing"]
+      "enum": ["success", "partial", "failed", "commit_missing", "docs_missing", "merge_conflict"]
     },
     "completed_phases": {
       "type": "array",

package/bundled/dev-pipeline/tests/conftest.py

@@ -1,133 +1,21 @@
-"""Shared fixtures for dev-pipeline tests."""
+"""Shared fixtures and import helpers for dev-pipeline tests."""
 
-import json
+import importlib
 import os
-import pytest
-
-
-@pytest.fixture
-def sample_feature_list():
-    """A valid feature-list.json structure."""
-    return {
-        "$schema": "dev-pipeline-feature-list-v1",
-        "app_name": "TestApp",
-        "global_context": {
-            "language": "Python",
-            "framework": "FastAPI",
-        },
-        "features": [
-            {
-                "id": "F-001",
-                "title": "Project Infrastructure Setup",
-                "description": "Set up the project structure.",
-                "priority": 1,
-                "estimated_complexity": "low",
-                "dependencies": [],
-                "acceptance_criteria": ["Has package.json", "Has README.md"],
-                "status": "pending",
-            },
-            {
-                "id": "F-002",
-                "title": "User Authentication",
-                "description": "Implement user login and registration.",
-                "priority": 2,
-                "estimated_complexity": "high",
-                "dependencies": ["F-001"],
-                "acceptance_criteria": ["Login works", "Registration works"],
-                "status": "pending",
-            },
-            {
-                "id": "F-003",
-                "title": "Dashboard View",
-                "description": "Create main dashboard.",
-                "priority": 3,
-                "estimated_complexity": "medium",
-                "dependencies": ["F-001", "F-002"],
-                "acceptance_criteria": ["Dashboard renders"],
-                "status": "pending",
-            },
-        ],
-    }
-
-
-@pytest.fixture
-def sample_bug_list():
-    """A valid bug-fix-list.json structure."""
-    return {
-        "$schema": "dev-pipeline-bug-fix-list-v1",
-        "project_name": "TestProject",
-        "global_context": {
-            "language": "Python",
-        },
-        "bugs": [
-            {
-                "id": "B-001",
-                "title": "Login crash on empty password",
-                "description": "App crashes when password is empty.",
-                "severity": "critical",
-                "priority": 1,
-                "error_source": {
-                    "type": "stack_trace",
-                    "stack_trace": "Traceback ... ValueError",
-                    "error_message": "Password cannot be empty",
-                },
-                "verification_type": "automated",
-                "acceptance_criteria": ["No crash on empty password"],
-                "status": "pending",
-                "affected_feature": "F-002",
-                "affected_modules": ["auth"],
-            },
-            {
-                "id": "B-002",
-                "title": "Slow dashboard loading",
-                "description": "Dashboard takes 10s to load.",
-                "severity": "medium",
-                "priority": 2,
-                "error_source": {
-                    "type": "user_report",
-                    "reproduction_steps": ["Open dashboard", "Wait 10 seconds"],
-                },
-                "verification_type": "manual",
-                "acceptance_criteria": ["Dashboard loads in <2s"],
-                "status": "pending",
-                "affected_feature": "F-003",
-                "environment": {"os": "macOS", "browser": "Chrome"},
-            },
-        ],
-    }
-
-
-@pytest.fixture
-def feature_list_file(tmp_path, sample_feature_list):
-    """Write sample feature list to a temp file and return the path."""
-    path = tmp_path / "feature-list.json"
-    path.write_text(json.dumps(sample_feature_list, indent=2), encoding="utf-8")
-    return str(path)
-
-
-@pytest.fixture
-def bug_list_file(tmp_path, sample_bug_list):
-    """Write sample bug list to a temp file and return the path."""
-    path = tmp_path / "bug-fix-list.json"
-    path.write_text(json.dumps(sample_bug_list, indent=2), encoding="utf-8")
-    return str(path)
-
-
-@pytest.fixture
-def state_dir(tmp_path):
-    """Create and return a basic state directory."""
-    sd = tmp_path / "state"
-    sd.mkdir()
-    features_dir = sd / "features"
-    features_dir.mkdir()
-    return str(sd)
-
-
-@pytest.fixture
-def bugfix_state_dir(tmp_path):
-    """Create and return a basic bugfix state directory."""
-    sd = tmp_path / "bugfix-state"
-    sd.mkdir()
-    bugs_dir = sd / "bugs"
-    bugs_dir.mkdir()
-    return str(sd)
+import sys
+
+# Add dev-pipeline/scripts to the import path so tests can import modules
+# without requiring package installation.
+_scripts_dir = os.path.join(os.path.dirname(__file__), "..", "scripts")
+sys.path.insert(0, _scripts_dir)
+
+# generate-bootstrap-prompt.py has a hyphenated filename that Python can't
+# import normally.  Load it via importlib and register under an underscore name
+# so test files can do: `from generate_bootstrap_prompt import ...`
+_spec = importlib.util.spec_from_file_location(
+    "generate_bootstrap_prompt",
+    os.path.join(_scripts_dir, "generate-bootstrap-prompt.py"),
+)
+_mod = importlib.util.module_from_spec(_spec)
+sys.modules["generate_bootstrap_prompt"] = _mod
+_spec.loader.exec_module(_mod)