Run any Skill in Manus with one click

spec-decomposer

Stars3

Forks1

UpdatedJune 30, 2026 at 20:43

Spec-Decomposer: splits monolithic framework specifications (FPF/DPF/LPF) into atomic skill reference files. Use when: decomposing a framework monolith into skill references, updating references/ after monolith changes, or setting up a new DPF/LPF skill from a monolith. Depends on fpf-core, dpf-fpf-literacy, dpf-lfw-architecture.

Installation

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

Run Skill in Manus

Source

dimonier

dimonier/skills

View GitHub Repository View Creator Repositories

Download

Run Skill in Manus

Related occupationsSOC

Based on SOC occupation classification

Computer Occupations, All OtherComputer and Mathematical Occupations·SOC 15-1299

File Explorer

28 files

SKILL.md

readonly

name	spec-decomposer
description	Spec-Decomposer: splits monolithic framework specifications (FPF/DPF/LPF) into atomic skill reference files. Use when: decomposing a framework monolith into skill references, updating references/ after monolith changes, or setting up a new DPF/LPF skill from a monolith. Depends on fpf-core, dpf-fpf-literacy, dpf-lfw-architecture.

DPF: Spec-Decomposer

Depends on: fpf-core, dpf-fpf-literacy, dpf-lfw-architecture Bounded context: Decomposing monolithic framework specs into LFW skills Source of truth: assets/SpecDecomposer-dpf.md

Quick Start: Decompose a Monolith

# For DPF/LPF monoliths (sections 1-10 structure):
python scripts/decompose_dpf.py <monolith.md> --skills-dir <output-skill-dir>

# For FPF monoliths (:End-delimited patterns):
python scripts/decompose_fpf.py <monolith.md> --skills-dir <output-skill-dir>

# Preview without writing:
python scripts/decompose_fpf.py <monolith.md> --skills-dir <output-skill-dir> --dry-run

The agent reads the monolith header, determines the type (FPF vs DPF), and invokes the correct script. Output is a complete skill: references/*.md + INDEX.md + (for DPF) relations.md.

When to load which pattern

Situation	Load	Governing cues
Unclear what to extract from monolith	`references/unit-recognition.md`	C.22.2, E.8, E.4.DPF
Need to determine problem card boundaries	`references/boundary-detection.md`	C.22.2, E.8
Need reference file format	`references/reference-file-format.md`	C.33, E.17.EFP, E.4.PFAD
Need to convert governing-pattern links	`references/governing-cue-conversion.md`	E.5.3, E.4.PFR
Need to generate INDEX.md	`references/index-generation.md`	C.33
Need to extract relation records	`references/relations-extraction.md`	E.4.PFR
Need to update SKILL.md after decomposition	`references/dispatcher-update.md`	EWA.SkillDispatcher
Agent tries to read monolith instead of references	`references/carrier-split.md`	EWA.MonolithInSkill
References out of sync with monolith	`references/sync-discipline.md`	EWA.MonolithInSkill, G.11
Need to decide what to automate	`references/automation-boundary.md`	E.16, G.2

Source for Agent vs Human

Agent: always use references/. Do NOT read assets/SpecDecomposer-dpf.md.
Human: read and edit the canonical monolith assets/SpecDecomposer-dpf.md. After edits, run scripts/decompose_dpf.py to rebuild references/.

References Status

Ready — 10 problem cards + INDEX + relations + scripts.

More from this repository

same repository

fpf-narrative

dimonier/skills

Generate compact, unambiguous FPF-structured output at F4-F5 formality level. Use when writing for AI agent consumers or FPF-literate humans where token economy and auditability matter: diagnostics, architecture decisions, code reviews, status reports, trust assessments, system compositions. Replaces verbose prose (F0) with typed-slot notation (~65% token savings, lossless). Do NOT use for: casual chat, teaching, non-technical audiences, creative tasks.

2026-07-113

fpf-core

dimonier/skills

First Principles Framework core patterns. Use when doing engineering, research, management, or mixed human/AI work. ALWAYS use as governing patterns for any DPF or LPF work. Load individual patterns from references/ as needed — never read the full spec unless doing deep audit.

2026-07-103

dpf-business-analysis

dimonier/skills

Business Analysis & Requirements Engineering. Use when: identifying stakeholders, eliciting or specifying requirements, prioritizing, validating, tracing, or managing requirements changes. Depends on fpf-core.

2026-06-283

dpf-fpf-literacy

dimonier/skills

FPF Literacy & DPF Authoring. Use when: helping an agent understand the FPF/DPF/LPF stack, creating a new DPF, loading domain knowledge into an agent, distinguishing FPF-grounded answers from generic AI responses, or choosing a carrier for publication. Depends on fpf-core.

2026-06-283

dpf-lfw-architecture

dimonier/skills

LFW Architecture. Use when: organizing the FPF/DPF/LPF workspace, deciding where to place a new framework, setting up the skill-carrier structure, or resolving LPF vs Project boundary questions. Depends on fpf-core.

2026-06-283

agent-skill-builder

dimonier/skills

Guide for deciding when to create agent skills, how to design them well, and what anti-patterns to avoid. Use when: (1) creating a new skill, (2) deciding between skills vs AGENTS.md vs MCP vs Memory, (3) reviewing an existing skill for quality, (4) unsure if a skill is needed at all.

2026-06-213

name	spec-decomposer
description	Spec-Decomposer: splits monolithic framework specifications (FPF/DPF/LPF) into atomic skill reference files. Use when: decomposing a framework monolith into skill references, updating references/ after monolith changes, or setting up a new DPF/LPF skill from a monolith. Depends on fpf-core, dpf-fpf-literacy, dpf-lfw-architecture.

DPF: Spec-Decomposer

Depends on: fpf-core, dpf-fpf-literacy, dpf-lfw-architecture Bounded context: Decomposing monolithic framework specs into LFW skills Source of truth: assets/SpecDecomposer-dpf.md

Quick Start: Decompose a Monolith

# For DPF/LPF monoliths (sections 1-10 structure):
python scripts/decompose_dpf.py <monolith.md> --skills-dir <output-skill-dir>

# For FPF monoliths (:End-delimited patterns):
python scripts/decompose_fpf.py <monolith.md> --skills-dir <output-skill-dir>

# Preview without writing:
python scripts/decompose_fpf.py <monolith.md> --skills-dir <output-skill-dir> --dry-run

The agent reads the monolith header, determines the type (FPF vs DPF), and invokes the correct script. Output is a complete skill: references/*.md + INDEX.md + (for DPF) relations.md.

When to load which pattern

Situation	Load	Governing cues
Unclear what to extract from monolith	`references/unit-recognition.md`	C.22.2, E.8, E.4.DPF
Need to determine problem card boundaries	`references/boundary-detection.md`	C.22.2, E.8
Need reference file format	`references/reference-file-format.md`	C.33, E.17.EFP, E.4.PFAD
Need to convert governing-pattern links	`references/governing-cue-conversion.md`	E.5.3, E.4.PFR
Need to generate INDEX.md	`references/index-generation.md`	C.33
Need to extract relation records	`references/relations-extraction.md`	E.4.PFR
Need to update SKILL.md after decomposition	`references/dispatcher-update.md`	EWA.SkillDispatcher
Agent tries to read monolith instead of references	`references/carrier-split.md`	EWA.MonolithInSkill
References out of sync with monolith	`references/sync-discipline.md`	EWA.MonolithInSkill, G.11
Need to decide what to automate	`references/automation-boundary.md`	E.16, G.2

Source for Agent vs Human

Agent: always use references/. Do NOT read assets/SpecDecomposer-dpf.md.
Human: read and edit the canonical monolith assets/SpecDecomposer-dpf.md. After edits, run scripts/decompose_dpf.py to rebuild references/.

References Status

Ready — 10 problem cards + INDEX + relations + scripts.