knowledge-graph

Three-Layer Memory System — automatic fact extraction, entity-based knowledge graph, and weekly synthesis. Manages life/areas/ entities with atomic facts and living summaries.

Install with Tessl CLI

npx tessl i github:jdrhyne/agent-skills --skill knowledge-graph

What are skills?

Review — 47%

Does it follow best practices?

If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:

npx tessl skill review --optimize ./path/to/skill

Learn more

Evaluation — 91%

Agent success when using this skill

Validation — 10 / 11 Passed

Validation for skill structure

SKILL.md

Review

Evals

Evaluation results

97%

-2%

Contact Intelligence Update

Fact extraction and entity management

Criteria

Without context

With context

Correct directory layout

100%

facts.jsonl present

100%

JSONL format

100%

Required fact fields

100%

Auto-increment ID format

100%

Atomic facts

87%

75%

Supersede old fact

100%

Preference supersession

100%

Ephemera skipped

100%

summary.md present

100%

summary.md format

100%

87%

New entity created

100%

Project entity created

100%

Without context: $0.3955 · 1m 21s · 18 turns · 19 in / 5,013 out tokens

With context: $0.8241 · 2m 24s · 32 turns · 32 in / 9,726 out tokens

79%

59%

New Agent Workspace Setup

Knowledge graph workspace initialization

Criteria

Without context

With context

People directory

Companies directory

Projects directory

AGENTS.md exists

100%

AGENTS.md KG block

50%

100%

AGENTS.md retrieval order

100%

AGENTS.md recall policy

100%

HEARTBEAT.md exists

100%

HEARTBEAT.md recall section

100%

Append-only rule in AGENTS.md

100%

New entity rule in AGENTS.md

12%

100%

Setup documentation

100%

Without context: $0.6689 · 2m 47s · 31 turns · 27 in / 8,432 out tokens

With context: $0.3537 · 1m 16s · 16 turns · 15 in / 4,015 out tokens

97%

Knowledge Base Maintenance Run

Weekly synthesis and summary generation

Criteria

Without context

With context

Tom summary updated

100%

Summary line count

100%

Last updated date

100%

Active facts only in summary

75%

100%

Meridian summary corrected

100%

Sofia contradiction resolved

50%

100%

Sofia summary updated

100%

Summary structure

100%

facts.jsonl append-only

70%

Synthesis report produced

100%

Report lists active fact counts

100%

Report lists supersessions

100%

Without context: $0.3067 · 1m 10s · 15 turns · 15 in / 4,082 out tokens

With context: $0.7843 · 2m 50s · 28 turns · 76 in / 10,613 out tokens

Evaluated: about 6 hours ago
Agent: Claude Code

Table of Contents

Contact Intelligence Update New Agent Workspace Setup Knowledge Base Maintenance Run

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.