skills.rest
NeoLabHQ
NeoLabHQOFFICIAL
Explore all skills by this author →
WebsiteGitHub

actualize

Reconciles FPF knowledge base state with recent git repository changes.

1.5k|154|Updated Nov 13, 2025
Search Tags:#evidence tracking#FPF#context drift#Git diff#epistemic debt#Knowledge Base#assurance case
One-click install
npx skills add https://github.com/NeoLabHQ/context-engineering-kit --skill actualize
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: actualize
Source: https://github.com/NeoLabHQ/context-engineering-kit/tree/main/plugins/fpf/skills/actualize
Command: npx skills add https://github.com/NeoLabHQ/context-engineering-kit --skill actualize

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

As a codebase evolves, the FPF knowledge base (.fpf/) drifts out of sync: evidence files reference changed code, decisions rest on stale foundations, and project context no longer matches reality. This Skill audits recent git changes against the knowledge base to surface context drift, stale evidence, and outdated decisions before they corrupt downstream reasoning.

Core Features & Use Cases

  • Context Drift Detection: Compares changed configuration files (package.json, Dockerfile, go.mod, etc.) against .fpf/context.md and prompts updates.
  • Evidence Staleness Audit: Cross-references carrier_ref fields in .fpf/evidence/ files with git diffs to flag stale evidence and affected hypotheses.
  • Decision Relevance Review: Traces DRR files in .fpf/decisions/ back to source evidence and flags potentially outdated decisions.
  • Baseline Tracking: Maintains a .fpf/.baseline file recording the last actualized commit for incremental audits.
  • Use Case: Before starting a new sprint, run the actualization audit to discover that three evidence files reference a refactored API handler, then re-validate those hypotheses before making architecture decisions.

Quick Start

Ask the agent to actualize the FPF knowledge base against recent git changes and report any stale evidence or outdated decisions.

Frequently Asked Questions about actualize

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

FAQPage Schema
How do I keep an FPF knowledge base in sync with code changes?▼

Run the actualize command, which diffs the current commit against the recorded baseline, checks changed files against context, evidence, and decision records, and reports drift. It then updates the .fpf/.baseline file with the new commit hash.

What is stale evidence in an assurance case?▼

Stale evidence is an evidence file whose carrier_ref points to a source file that changed since the evidence was recorded. The audit flags these items and identifies which hypotheses they support so you can re-validate them.

When should I run the actualization audit?▼

Run it before starting new work, after major code changes, as part of weekly maintenance, and before making decisions that depend on existing evidence. This ensures the knowledge base reflects current repository reality.

Does the actualize command modify evidence or decision files automatically?▼

No, it only flags stale evidence and potentially outdated decisions in a structured report. Updating context.md, re-validating evidence, or re-evaluating decisions requires explicit user confirmation or follow-up commands.

What happens if no baseline file exists yet?▼

If .fpf/.baseline does not exist, the audit uses the repository's initial commit as the comparison point. After the run, it creates the baseline file recording the current commit, timestamp, and branch.

Related Skills

Explore More Software Engineering Skills→
luicabref97luicabref97

fpf:query

Search and aggregate FPF knowledge base hypotheses with assurance scores.

Community
Intermediate
luicabref97luicabref97

fpf:status

Display FPF knowledge base status with hypothesis counts, evidence freshness, and decision records.

Community
Intermediate
arbgjrarbgjr

decay-scoring

Compute obsolescence scores for RAG corpus knowledge nodes using configurable weights.

Community
Advanced
paulinett1508-devpaulinett1508-dev

audit-dead-code-auditor-references-checklist

Loads a YAML-described skill entrypoint and produces category-based dead-code review guidance from canonical references.

Community
Basic