mcpField
A
agently-session-memory

agently-session-memory

github.com/agenteraAgent workflows3140 installsUpdated 4d ago
Repository

About agently-session-memory

Agently Session Memory is a skill for designing continuity and restore behavior in an existing request flow. It recommends Session-backed memory, explicit chat-history and memo boundaries, and separating request memory from workflow state.

Key features

  1. 01Session-backed continuity guidance
  2. 02Explicit chat-history and memo boundaries
  3. 03Separation of request memory and workflow state
  4. 04Avoidance of transient restart-sensitive memory

Use cases

  1. 01Adding continuity to an existing request flow
  2. 02Designing memory that survives restarts

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/agentera/agently-skills --skill agently-session-memory

After installing, type/agently-session-memoryin Claude Code to confirm it loaded.

Bundled files

References and scripts installed along with SKILL.md.

examples/
.gitkeep
outputs/
.gitkeep
references/
overview.md
scripts/
.gitkeep

Details

From the project README

Agently Session Memory

Use this skill when the request path already exists but continuity or restore behavior is the main design problem.

Native-First Rules

  • prefer Session-backed continuity before inventing a custom memory layer
  • keep request-side chat history and memo boundaries explicit
  • separate request memory from workflow runtime state

Anti-Patterns

  • do not use session as a substitute for workflow orchestration state
  • do not keep restart-sensitive memory only in transient globals

Read Next

  • references/overview.md

Comments

Markdown supported