skills.rest
catlog22
catlog22COMMUNITY
Explore all skills by this author →
GitHub

spec-map

Analyze codebases to generate mapper outputs for tech stack, architecture, features, and concerns.

511|63|Updated Mar 17, 2026
Search Tags:#automation#codebase#analysis#multi-agent#orchestration#csv#mapping
One-click install
npx skills add https://github.com/catlog22/Maestro-Flow --skill spec-map
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: spec-map
Source: https://github.com/catlog22/Maestro-Flow/tree/main/.codex/skills/spec-map
Command: npx skills add https://github.com/catlog22/Maestro-Flow --skill spec-map

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Maestro Spec Map removes manual, piecemeal codebase reconnaissance by running four focused mapper agents in parallel to produce concise, discoverable documentation about tech stack, architecture, features, and cross-cutting concerns.

Core Features & Use Cases

  • Single-wave parallel mapping: Launches four independent mappers (tech stack, architecture, features, concerns) concurrently to maximize throughput.
  • CSV-driven orchestration: Uses a tasks.csv source of truth and spawn_agents_on_csv to reliably track state, results, and partial failures.
  • Practical outputs: Aggregates findings into .workflow/codebase/ files and a context summary, useful for onboarding, audits, and architecture reviews.

Quick Start

Run the spec-map skill with an optional focus and flags to auto-confirm and set concurrency, for example: $spec-map "auth" -y -c 4

Frequently Asked Questions about spec-map

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate codebase analysis to generate architecture and tech stack documentation?▼

Parallel codebase mapping uses concurrent agents to analyze tech stack, architecture, features, and concerns simultaneously. It uses a CSV-driven orchestration model to reliably track state, results, and partial failures during single-wave repository reconnaissance.

Can I scope codebase mapping to a specific feature and auto-confirm the process?▼

Yes, codebase mapping supports optional focus scoping and auto-confirm flags. You can target specific areas like "auth" and set concurrency levels to rapidly generate scoped mapper outputs without manual confirmation prompts.

Does multi-agent codebase orchestration work with repositories of any size?▼

Codebase mapping outputs are aggregated into .workflow/codebase/ files and a context summary. Discoveries are collected in NDJSON format, providing concise documentation useful for onboarding, audits, and architecture reviews.

What file formats are used for orchestrating parallel codebase mapping tasks?▼

Parallel codebase mapping uses a tasks.csv file as the source of truth for orchestration. Agent discoveries are collected in NDJSON format, and aggregated summaries are written as files into the .workflow/codebase/ directory.

What's the best way to map cross-cutting concerns and features across a large codebase?▼

The best way to map cross-cutting concerns is using single-wave parallel mapping to concurrently analyze the repository. This removes manual reconnaissance by spawning independent agents that generate structured documentation for features and concerns.

Related Skills

Explore More Software Engineering Skills→
catlog22catlog22
511

maestro-execute

Execute plan.json tasks in parallel waves with CSV-driven state.

Community
Advanced
dakshrawat298-gifdakshrawat298-gif

security-audit-context-building

Map architecture, trust boundaries, data flows, and threat models into an Audit Context Document.

Community
Advanced
catlog22catlog22
511

quality-test-gen

Generate parallel unit, integration, and e2e test suites from a CSV of source files.

Community
Advanced
bhanukushwahbhanukushwah

map-codebase

Orchestrate parallel agents to analyze codebases and write structured documents.

Community
Advanced