opencontext

Persistent memory and context management for AI agents using OpenContext. Keep context across sessions/repos/dates, store conclusions, and provide document search workflows.

Install with Tessl CLI

npx tessl i github:supercent-io/skills-template --skill opencontext

What are skills?

2.13x

Quality

55%

Does it follow best practices?

Impact

96%

2.13x

Average score across 3 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./.agent-skills/opencontext/SKILL.md

SKILL.md

Review

Evals

Evaluation results

95%

87%

Setting Up Persistent AI Memory for a Development Team

Project init and daily workflow

Criteria

Without context

With context

CLI install command

100%

oc init in script

100%

Before-work slash command

100%

During-work slash command

100%

After-work slash command

100%

Three-phase workflow present

100%

Contexts storage path

100%

Database path

100%

Acceptance Criteria doc type

100%

Common Pitfalls doc type

100%

API Contracts doc type

100%

Dependency Versions doc type

100%

AGENTS.md update mentioned

Without context: $0.2273 · 1m 15s · 11 turns · 14 in / 3,630 out tokens

With context: $0.3607 · 1m 14s · 20 turns · 207 in / 3,658 out tokens

93%

Automating Context Retrieval for a Research Team

Search modes and document management

Criteria

Without context

With context

Folder create command

100%

Doc create command

100%

Search command syntax

100%

Keyword mode flag

100%

EMBEDDING_API_KEY config

100%

Index build step

100%

Manifest command

100%

Manifest limit flag

100%

Keyword mode: no embeddings

100%

Vector/hybrid mode requirements

100%

Hybrid as default

100%

Optional embedding config keys

Without context: $0.9110 · 5m 15s · 47 turns · 45 in / 9,921 out tokens

With context: $0.4079 · 1m 25s · 21 turns · 26 in / 4,559 out tokens

100%

66%

Designing a Multi-Agent Development Pipeline with Persistent Knowledge

Stable links and multi-agent workflow

Criteria

Without context

With context

Claude as planner first

57%

100%

Gemini as analyst second

71%

100%

Claude as coder third

57%

100%

Codex for run/test fourth

71%

100%

Claude synthesizes and stores last

100%

All six MCP tools listed

100%

Stable link format

13%

100%

CLI link generation command

25%

100%

MCP link generation call

16%

100%

OpenContext search in analysis phase

42%

100%

Results stored after final phase

85%

100%

Without context: $1.0317 · 3m 35s · 28 turns · 30 in / 7,725 out tokens

With context: $0.3587 · 1m 18s · 20 turns · 273 in / 3,878 out tokens

Evaluated: 3 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Setting Up Persistent AI Memory for a Development Team Automating Context Retrieval for a Research Team Designing a Multi-Agent Development Pipeline with Persistent Knowledge

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.