skills.rest
egregore-labs
egregore-labsOFFICIAL
Explore all skills by this author →
Website𝕏@egregore_xyzGitHub

deep-reflect

Cross-reference insights against a git-backed knowledge graph to detect structural signals.

275|17|Updated Mar 24, 2026
Search Tags:#analysis#signals#reflection#artifact#knowledge-graph#neo4j#deep-reflect
One-click install
npx skills add https://github.com/egregore-labs/egregore --skill deep-reflect
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deep-reflect
Source: https://github.com/egregore-labs/egregore/tree/main/.claude/skills/deep-reflect
Command: npx skills add https://github.com/egregore-labs/egregore --skill deep-reflect

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Deep-reflect turns ad-hoc insights and findings into evidence-backed reflections by automatically retrieving relevant artifacts, running iterative multi-sample analyses, and surfacing structurally meaningful signals (tensions, gaps, convergences, dependencies) so teams can make decisions with grounded context instead of intuition.

Core Features & Use Cases

  • Evidence-driven cross-referencing: Retrieves candidate artifacts from the organization's memory and inspects file content to justify each signal with excerpts and provenance.
  • Multi-mode analysis: Supports deep (multi-sample consensus), focused (topic-filtered), and quick (single-pass) modes to trade off latency and depth.
  • Signal ontology & graph integration: Classifies relationships (tension, convergence, gap, dependency, etc.), writes a reflective artifact to memory, and creates graph edges to record reasoning and consequences.
  • Use Case: Use after a design decision or incident to understand whether the insight contradicts prior decisions, uncovers missing prerequisites, or reinforces an existing pattern.

Quick Start

Ask the assistant to "deep-reflect 'finding: Neo4j queries are slow'" to run an evidence-based cross-reference against the team's memory.

Frequently Asked Questions about deep-reflect

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

FAQPage Schema
How do I cross-reference a project insight against my team's knowledge base?▼

Cross-reference a project insight by retrieving relevant artifacts from a Neo4j-backed memory, running multi-sample analyses, and surfacing structural signals like tensions, gaps, or convergences with excerpted evidence.

What is evidence-based reflection in organizational knowledge graphs?▼

Evidence-based reflection in organizational knowledge graphs is the process of validating ad-hoc findings against stored artifacts, classifying relationships like dependencies or redundancies, and persisting a reflective artifact with provenance back into memory.

How do I detect decision tensions or gaps in a Neo4j knowledge graph?▼

Detect decision tensions or gaps by applying a signal ontology that evaluates artifact contents, classifies structural convergences or dependencies, and writes graph edges to record reasoning and consequences within the Neo4j memory.

Can I run focused or quick analysis modes to reduce knowledge graph cross-referencing latency?▼

You can run focused or quick analysis modes to reduce knowledge graph cross-referencing latency, trading multi-sample consensus depth for a single-pass or topic-filtered evaluation of retrieved artifacts.

Do I need Neo4j to evaluate findings and surface structurally significant signals?▼

You need a Neo4j-backed memory to evaluate findings and surface structurally significant signals, as the process requires querying the graph, accessing artifact file contents for evidence, and persisting reflective artifacts.

When should I use deep analysis over quick single-pass reflection on artifacts?▼

Use deep analysis over quick single-pass reflection when evaluating complex design decisions or incidents, requiring multi-sample consensus to uncover missing prerequisites and grounded context instead of relying on intuition.

Related Skills

Explore More Product & Management Skills→
RoentekRoentek
4

lightrag

Build graph-based retrieval augmented generation over documents for entity-aware Q&A.

Official
Advanced
Chris-MaskeyChris-Maskey

graph-thinking

Constructs graph models of nodes, edges, and clusters to analyze complex systems and relationships.

Community
Advanced
stephen94125stephen94125
2

read-file

Retrieve file contents from the filesystem and return them as a string.

Community
Basic
robwestzrobwestz

search

Perform semantic search across MemPalace with knowledge graph cross-referencing.

Community
Intermediate