smithery/mehdic

config-seeder

Seed JSON configuration files into database. Use ONCE at BAZINGA session initialization, BEFORE spawning PM.

Installation

$ npx skills add smithery/mehdic --skill config-seeder

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from smithery/mehdic · top by installs.

npx skills add smithery/mehdic

Browse all from smithery/mehdic

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Skill metadata

Parsed from SKILL.md frontmatter.

Version1.0.0
Allowed toolsBash
Declared agents claude-code

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 3,717 B
  • docs SUMMARY.md 129 B

History

  1. First recorded snapshot · 0 installs

SKILL.md

Config Seeder Skill

You are the config-seeder skill. Your role is to seed JSON configuration files into database tables at session start.

Overview

This skill seeds workflow configuration from JSON files into database tables:

  • workflow/transitions.json → workflow_transitions table
  • workflow/agent-markers.json → agent_markers table
  • specialrules from transitions.json → workflowspecialrules table

Note: In dev mode, bazinga/config is a symlink to ../workflow for path compatibility.

Prerequisites

  • Database must be initialized (bazinga/bazinga.db exists)
  • Config files must exist:

- workflow/transitions.json (source) or bazinga/config/transitions.json (symlink) - workflow/agent-markers.json (source) or bazinga/config/agent-markers.json (symlink)

When to Invoke This Skill

  • ONCE at session start, BEFORE spawning PM
  • Part of session initialization (after create-session, before PM spawn)
  • Should NOT be invoked during normal orchestration flow

Your Task

When invoked, you must:

Step 1: Call the Session Init Script (RECOMMENDED)

Use the unified initialization script that handles DB + config in one step:

python3 .claude/skills/bazinga-db/scripts/init_session.py --session-id "{session_id}"

This script:

  1. Ensures database exists with correct schema
  2. Seeds workflow configs (transitions, markers, rules)
  3. Creates artifacts directory
  4. Verifies everything is ready

Alternative: Call seed script directly (requires DB to exist first):

python3 .claude/skills/config-seeder/scripts/seed_configs.py --all

With auto-init: Initialize DB if missing:

python3 .claude/skills/config-seeder/scripts/seed_configs.py --all --auto-init-db

Step 2: Verify Success

Expected output on success:

Seeded 45 transitions
Seeded 7 agent marker sets
Seeded 5 special rules
✅ Config seeding complete

Step 3: Report Result

If successful:

  • Report: "Configuration seeded successfully"
  • Orchestration can proceed

If failed:

  • Report the error message
  • Orchestration CANNOT proceed without routing config

What Gets Seeded

Transitions (45+ entries)

State machine rules for routing:

  • developer + READYFORQA → qa_expert
  • developer + READYFORREVIEW → tech_lead
  • qaexpert + PASS → techlead
  • tech_lead + APPROVED → merge → check phase
  • etc.

Agent Markers (7 entries)

Required markers per agent type:

  • developer: "NO DELEGATION", "READYFORQA", "BLOCKED", etc.
  • qa_expert: "PASS", "FAIL", "Challenge Level", etc.
  • techlead: "APPROVED", "CHANGESREQUESTED", etc.
  • etc.

Special Rules (5 entries)

Workflow modification rules:

  • testingmodedisabled: Skip QA entirely
  • testingmodeminimal: Skip QA Expert
  • escalationafterfailures: Escalate to SSE after 2 failures
  • security_sensitive: Force SSE + mandatory TL review
  • research_tasks: Route to RE with limited parallelism

Output Format

Return success/failure message. No structured output needed.

Error Handling

Error Meaning
Config file not found JSON files missing - check workflow/ (or bazinga/config/ symlink)
Database not found Run init_db.py first
Insert failed Database schema mismatch - run migration

If seeding fails, orchestration CANNOT proceed. The prompt-builder and workflow-router skills depend on these database tables.