How do I install the Pm DB skill?

Run npx skillmds@latest add artsmc-claude-dev-agents/pm-db in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Pm DB safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: executes scripts. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Pm DB?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Pm DB free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

artsmc (@artsmc-claude-dev-agents) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Pm DB by artsmc-claude-dev-agents — Agent Skill

Name: Pm DB
Rating: 4.5 (4 reviews)
Author: artsmc

Pm DB

Project management database for tracking specs, phases, tasks, and execution runs. Use this skill whenever the user asks about project status, wants to track feature work, needs a dashboard of progress, wants to import specs from /spec-plan, or mentions tracking tasks, phases, or runs. Also use when the user says things like 'what's the status', 'how's the project going', 'track this work', 'show me progress', or 'import my specs'.

artsmc Updated Sep 10, 2026

File contents

PM-DB: Project Management Database

SQLite-based tracking for the full development lifecycle: specs → phases → tasks → runs.

Core Concepts

PM-DB tracks work in a hierarchy:

Project (e.g., "acme")
  └─ Phase (e.g., "file-upload-feature")
       └─ Plan (approved task breakdown)
            └─ Tasks (atomic work items with waves/dependencies)
                 └─ Runs (execution attempts with timing/status)

Why this matters: When a user asks "what's the status?", you query projects → phases → tasks. When they say "track this feature", you create a phase with a plan and tasks.

Commands & Import Walkthrough

Full command invocations (init, import, dashboard, migrate, export) and the step-by-step import walkthroughs (including manual Python-API import) live in references/commands.md — read it before running any pm-db script or importing specs.

Cross-Skill Integration

PM-DB connects to other skills in the workflow:

/spec-plan → generates specs in job-queue/
     ↓
/pm-db import → imports specs into projects/phases
     ↓
/start-phase-plan → creates execution plan with tasks
     ↓
/pm-db tracks → phases, runs, task completions
     ↓
/pm-db dashboard → shows progress

Key paths:

Spec-plan output: /home/artsmc/applications/acme/job-queue/feature-*/docs/
PM-DB database: ~/.claude/projects.db
Python API: ~/.claude/lib/project_database.py

Python API (Quick Reference)

from project_database import ProjectDatabase

with ProjectDatabase() as db:
    # Projects
    projects = db.list_projects()
    project = db.get_project_by_name("acme")

    # Phases
    phases = db.list_phases(project_id=1)
    phase = db.get_phase(phase_id=1)

    # Plans and tasks
    plans = db.list_phase_plans(phase_id=1)
    tasks = db.list_tasks(plan_id=1)

    # Runs
    run_id = db.create_phase_run(phase_id=1, plan_id=1)
    db.start_phase_run(run_id)
    db.complete_phase_run(run_id, status='completed')

    # Dashboard
    dashboard = db.generate_phase_dashboard(phase_id=1)
    metrics = db.get_phase_metrics(phase_id=1)

For the complete API, read ~/.claude/lib/project_database.py.

Database Schema (v5)

Core tables: projects, phases, phase_plans, plan_documents, tasks, task_dependencies, phase_runs, task_runs

Tracking tables: task_updates, quality_gates, code_reviews, run_artifacts, phase_metrics

Advanced: agent_context_cache, agent_invocations, agent_file_reads, checklists, checklist_items

Troubleshooting

"Database not found" → Run python3 ~/.claude/skills/pm-db/scripts/init_db.py

"No projects found" → Run import: python3 ~/.claude/skills/pm-db/scripts/import_specs.py --auto-confirm

"Import found nothing" → Check the job-queue path. Default is ~/.claude/job-queue/. For the monorepo, add --job-queue-dir /home/artsmc/applications/acme/job-queue

"Which import script?" → import_specs.py for spec files, import_phases.py for structured phases

"Database locked" → Check lsof ~/.claude/projects.db. WAL mode is enabled by default.

Database: ~/.claude/projects.db API: ~/.claude/lib/project_database.py Version: 2.0

---name: pm-dbdescription: Project management database for tracking specs, phases, tasks, and execution runs. Use this skill whenever the user asks about project status, wants to track feature work, needs a dashboard of progress, wants to import specs from /spec-plan, or mentions tracking tasks, phases, or runs. Also use when the user says things like 'what's the status', 'how's the project going', 'track this work', 'show me progress', or 'import my specs'.---# PM-DB: Project Management DatabaseSQLite-based tracking for the full development lifecycle: specs → phases → tasks → runs.## Core ConceptsPM-DB tracks work in a hierarchy:```Project (e.g., "acme") └─ Phase (e.g., "file-upload-feature") └─ Plan (approved task breakdown) └─ Tasks (atomic work items with waves/dependencies) └─ Runs (execution attempts with timing/status)```**Why this matters:** When a user asks "what's the status?", you query projects → phases → tasks. When they say "track this feature", you create a phase with a plan and tasks.## Commands & Import WalkthroughFull command invocations (init, import, dashboard, migrate, export) and the step-by-step import walkthroughs (including manual Python-API import) live in `references/commands.md` — read it before running any pm-db script or importing specs.## Cross-Skill IntegrationPM-DB connects to other skills in the workflow:```/spec-plan → generates specs in job-queue/ ↓/pm-db import → imports specs into projects/phases ↓/start-phase-plan → creates execution plan with tasks ↓/pm-db tracks → phases, runs, task completions ↓/pm-db dashboard → shows progress```**Key paths:**- Spec-plan output: `/home/artsmc/applications/acme/job-queue/feature-*/docs/`- PM-DB database: `~/.claude/projects.db`- Python API: `~/.claude/lib/project_database.py`## Python API (Quick Reference)```pythonfrom project_database import ProjectDatabasewith ProjectDatabase() as db: # Projects projects = db.list_projects() project = db.get_project_by_name("acme") # Phases phases = db.list_phases(project_id=1) phase = db.get_phase(phase_id=1) # Plans and tasks plans = db.list_phase_plans(phase_id=1) tasks = db.list_tasks(plan_id=1) # Runs run_id = db.create_phase_run(phase_id=1, plan_id=1) db.start_phase_run(run_id) db.complete_phase_run(run_id, status='completed') # Dashboard dashboard = db.generate_phase_dashboard(phase_id=1) metrics = db.get_phase_metrics(phase_id=1)```For the complete API, read `~/.claude/lib/project_database.py`.## Database Schema (v5)Core tables: `projects`, `phases`, `phase_plans`, `plan_documents`, `tasks`, `task_dependencies`, `phase_runs`, `task_runs`Tracking tables: `task_updates`, `quality_gates`, `code_reviews`, `run_artifacts`, `phase_metrics`Advanced: `agent_context_cache`, `agent_invocations`, `agent_file_reads`, `checklists`, `checklist_items`## Troubleshooting**"Database not found"** → Run `python3 ~/.claude/skills/pm-db/scripts/init_db.py`**"No projects found"** → Run import: `python3 ~/.claude/skills/pm-db/scripts/import_specs.py --auto-confirm`**"Import found nothing"** → Check the job-queue path. Default is `~/.claude/job-queue/`. For the monorepo, add `--job-queue-dir /home/artsmc/applications/acme/job-queue`**"Which import script?"** → `import_specs.py` for spec files, `import_phases.py` for structured phases**"Database locked"** → Check `lsof ~/.claude/projects.db`. WAL mode is enabled by default.---**Database:** `~/.claude/projects.db`**API:** `~/.claude/lib/project_database.py`**Version:** 2.0

Frequently asked questions