Skip to content
Back to skills

Agent Session Memory

ASecurity

"Guides Cognee session memory, typed memory entries, feedback,

  • 247 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 8, 2026
ai-agentsapidatabasebackend

Works with

  • cli
  • api
  • mcp

Security analysis

A100/100

Pro scans all 5 files and shows the line behind each finding

Scanned September 8, 2026

npx -y skills add VectorSpaceLab/AREX-Skill --skill agent-session-memory --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Agent Session Memory?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Agent Session Memory
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/vectorspacelab-agent-session-memory/badge)](https://www.skillsdirectory.com/skills/vectorspacelab-agent-session-memory)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: "agent-session-memory"
description: "Guides Cognee session memory, typed memory entries, feedback,
  agent isolation, and agent-memory decorator workflows."
disable-model-invocation: true
metadata:
  disco-role: operating
license: Apache 2.0
---

# Cognee Agent and Session Memory

Use this sub-skill when the user wants Cognee to remember conversation state, store agent traces, record feedback, isolate memory per agent, or decorate an async agent with memory retrieval and persistence.

## Route here for

- `QAEntry`, `TraceEntry`, `FeedbackEntry`, and `SkillRunEntry` payloads.
- `recall(..., session_id=..., scope=...)` behavior.
- `@cognee.agent_memory(...)` decorator configuration.
- `cognee.agents.create/list/get/delete/register/unregister/list_connections/get_connection`.
- Session and feedback APIs: `get_session`, `add_feedback`, `delete_feedback`, frequency weights.
- Session distillation, feedback-influenced retrieval, and multi-user/agent isolation.

## Route away

- Basic permanent-memory storage and recall: [core-memory](../core-memory/SKILL.md).
- Retrieval mode tuning: [search-retrieval](../search-retrieval/SKILL.md).
- Backend/cache/env configuration: [configuration-backends](../configuration-backends/SKILL.md).
- CLI/MCP/UI service deployment: [api-cli-services](../api-cli-services/SKILL.md).

## Read first

1. [references/agent-session-memory.md](references/agent-session-memory.md)
2. [references/api-reference.md](references/api-reference.md)
3. [references/troubleshooting.md](references/troubleshooting.md)

## Safe helper

- [scripts/check_agent_memory_payloads.py](scripts/check_agent_memory_payloads.py) — validates typed memory payload JSON without writing to a database.

## Working rules

- Use a dedicated `session_id` per agent or task to avoid mixing unrelated traces.
- For feedback updates, keep the `qa_id` returned by a prior QA memory entry.
- Decorate only async functions with `@cognee.agent_memory`.
- Resolve dataset access before granting agent permissions.
- Route cache/provider failures to configuration instead of treating them as agent API problems.

Files in this skill

  • SKILL.md2.1 KB
  • references/agent-session-memory.md3.1 KB
  • references/api-reference.md2.8 KB
  • references/troubleshooting.md1.9 KB
  • scripts/check_agent_memory_payloads.py1.7 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.