salve

Extract session outcomes and update markdown memory files with a timestamped log.

520|175|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/evolution-foundation/evo-nexus --skill salve
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: salve
Source: https://github.com/evolution-foundation/evo-nexus/tree/main/.claude/skills/salve
Command: npx skills add https://github.com/evolution-foundation/evo-nexus --skill salve

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Keeps an agent session coherent by turning what happened during a workflow into updated memory files and a concise, timestamped session log.

Core Features & Use Cases

  • Session review to extract key outcomes: identifies decisions, created/resolved pending items, mentioned people with updated context, project updates, relevant metrics, and deadlines.
  • Targeted memory propagation: updates decisions, pendencies, people context, and per-project memory when a project was discussed, and checks for cross-references.
  • Session logging for traceability: writes a structured log to memory/sessions/YYYY-MM-DD.md so future runs can reference prior context.

Quick Start

Invoke salve after you finish a multi-agent session to have it review the session and update the corresponding memory context files and session log automatically.

Frequently Asked Questions about salve

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

FAQPage Schema
How do I save agent session outcomes to persistent markdown memory?

To save agent session outcomes to persistent markdown memory, invoke a session flushing tool after your workflow. It extracts decisions, pending items, and project updates, then writes them into specific memory context markdown files for durable storage.

How does decision tracking work for multi-agent workflows?

Decision tracking for multi-agent workflows works by reviewing the completed session to identify key choices and context. It then propagates these decisions and people context into targeted memory markdown files, ensuring future runs can reference past determinations.

Can I generate a timestamped session log for my knowledge base?

Yes, you can generate a timestamped session log for your knowledge base. The process creates a structured markdown record at memory/sessions/YYYY-MM-DD.md, capturing session traceability for future reference and synchronization.

What is the best way to synchronize markdown memory files with an external file watcher?

The best way to synchronize markdown memory files with an external file watcher is to write structured session logs into a standardized date-based markdown path. This allows the external Brain Repo file watcher to automatically detect and sync memory updates.

Does this session logging approach require specific project folder structures?

Yes, this session logging approach requires specific project folder structures. You must have memory/context and memory/projects markdown files, along with a memory/sessions directory, to correctly store extracted updates and timestamped logs.

What is included when flushing a session into memory stores?

Flushing a session into memory stores includes extracting decisions, pendencies, people context, project updates, relevant metrics, and deadlines. It checks for cross-references and updates corresponding markdown files to maintain session coherence.