skills.rest
TDMorales
TDMoralesCOMMUNITY
Explore all skills by this author →
GitHub

extracted-fact-management

Manage a serializable fact ledger across multi-step implementation tasks.

Updated Jan 23, 2026
Search Tags:#state management#checkpointing#persistence#agentic workflow#fact extraction#interruption handling
One-click install
npx skills add https://github.com/TDMorales/skill_set_library --skill extracted-fact-management
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: extracted-fact-management
Source: https://github.com/TDMorales/skill_set_library/tree/main/skills/.experimental/extracted-fact-management
Command: npx skills add https://github.com/TDMorales/skill_set_library --skill extracted-fact-management

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill ensures that crucial information extracted during a multi-step task is systematically recorded, preserved, and retrievable, preventing data loss due to interruptions, context switches, or session gaps.

Core Features & Use Cases

  • Fact Ledger: Maintains a structured, serializable record of all extracted facts, including their source, content, and status (PENDING, CONSUMED, STALE).
  • Checkpointing: Creates named snapshots of the ledger at the completion of each implementation step, serving as re-entry points for task resumption.
  • Interruption Handling: Manages SOFT, CLARIFY, REDIRECT, and GATE interruptions, ensuring the ledger state is preserved and the task can resume correctly.
  • Use Case: When an agent is implementing a complex feature that spans multiple files and sessions, this skill ensures that all extracted requirements, configurations, and intermediate data points are saved, so the agent can reliably resume work even after being interrupted by user questions or a change in direction.

Quick Start

Use the extracted-fact-management skill to manage facts during a multi-step implementation task.

Frequently Asked Questions about extracted-fact-management

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

FAQPage Schema
How do I preserve extracted facts during agentic workflow interruptions?▼

To preserve extracted facts during agentic workflow interruptions, use a structured fact ledger that records fact sources, content, and status, enabling reliable resumption after session gaps or context switches.

What is checkpointing in multi-step state management for agents?▼

Checkpointing in multi-step state management creates named snapshots of the fact ledger upon completing each implementation step, serving as stable re-entry points for task resumption after interruptions.

How do I handle SOFT and GATE interruptions when managing implementation state?▼

Handling SOFT, CLARIFY, REDIRECT, and GATE interruptions requires preserving the ledger state so the task can resume correctly, ensuring no extracted requirements or intermediate data are lost.

Can I resume a multi-session feature implementation without losing intermediate data points?▼

Yes, you can resume a multi-session feature implementation without losing intermediate data points by maintaining a serializable checkpoint ledger that tracks PENDING, CONSUMED, and STALE fact statuses.

What's the best way to manage fact extraction across files during complex implementation tasks?▼

The best way to manage fact extraction across complex tasks is integrating Lean Markdown Reading with a structured ledger, ensuring data integrity via hard invariants and staleness rules.

When should I not use a serializable fact ledger for state management?▼

You should avoid using a serializable fact ledger for state management if your task is single-step, requires no resumption across interruptions, or involves no multi-session fact extraction.

Related Skills

Explore More Software Engineering Skills→
ANcpLuaANcpLua
5

epistemic-checkpoint

Verify software metadata facts via local checks and official web sources.

Community
Intermediate
acaprinoacaprino
6

executing-plans

Execute implementation plans with verification checkpoints in separate sessions.

Community
Intermediate
robwestzrobwestz

heavy-file-ingestion-codex

Convert heavyweight PDFs, Office documents, and spreadsheets into markdown or CSV artifacts.

Community
Intermediate
SterlingChinSterlingChin
1.0k

update

Append session progress to sessions/{TODAY}.md and update state/current.md.

Community
Basic