memory-systems
An agent skill by muratcankoylan, from muratcankoylan/Agent-Skills-for-Context-Engineering. Tags: analytics, api, automation, developer-tools, strategy.
What it does
This skill should be used for persistent semantic memory in agent systems: cross-session knowledge retention, entity tracking, temporal validity, graph or vector retrieval, memory consolidation, and memory benchmark selection. Route file-backed scratchpads to filesystem-context, handoff summaries to context-compression, and token-efficiency tactics to context-optimization.
Install
With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:
npx skills add muratcankoylan/Agent-Skills-for-Context-Engineering --skill memory-systemsOr copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):
git clone --depth 1 https://github.com/muratcankoylan/Agent-Skills-for-Context-Engineering
cp -r Agent-Skills-for-Context-Engineering/skills/memory-systems ~/.claude/skills/memory-systemsSafety box score
Grade A, minimal risk: 9 of 100 risk points, rated 2026-10-20. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.
This skill is a mostly read-only design/reference package for agent memory systems, plus one local Python demo module. The main risks are limited to local code execution if the script is run and a small amount of prompt-instruction content about system-prompt placement; it does not appear to access files, networks, secrets, or elevated privileges.
| Category | Level | Why |
|---|---|---|
| Code & shell execution | 1 (low) | `scripts/memory_store.py` is an executable Python module with a `__main__` smoke test, so installing and running it will evaluate local code; the behavior is in-memory and demo-only (`if __name__ == "__main__":`). |
| Network access | 0 (none) | No code in the shipped files makes network calls, downloads content, or sends data off-machine; the external items are only references in markdown. |
| Filesystem reach | 0 (none) | The code is in-memory only and does not read, write, or delete files outside the project; there are no file paths, open/write calls, or shell redirections. |
| Secrets & credentials | 0 (none) | There is no handling of tokens, keychains, `.env` files, browser sessions, or credential prompts; examples only store ordinary text facts. |
| Privilege & persistence | 0 (none) | Nothing requires sudo, global installs, services, scheduled tasks, or shell/OS configuration changes. |
| Instruction hijack surface | 1 (low) | The skill text includes prompt-policy language such as `Scratchpad in system prompt` and `Always — optimize with attention-favored positions`, which is instruction-like content about how the agent should place memory in context, but it does not tell the agent to ignore safeguards or fetch external instructions. |
| Irreversible actions | 0 (none) | The skill only describes memory retrieval and consolidation; it does not push, post, pay, deploy, merge, or delete anything. |
| Transparency | 0 (none) | All shipped content is plain markdown and Python; there are no binaries, minified blobs, base64 payloads, or unpinned installer downloads. |
Before installing
- Read `scripts/memory_store.py` if you want to confirm the demo is in-memory only and has no file/network I/O.
- Check whether your agent policy allows skills that include prompt-placement instructions like `Scratchpad in system prompt`.
- If you extend the skill, inspect any added backends for file writes, network calls, or secret handling before enabling them.
Rated 2026-10-20 by an automated review of 3 files from the skill's repository.
Source
- Repository
- muratcankoylan/Agent-Skills-for-Context-Engineering (all skills from this repository)
- Path
- skills/memory-systems/SKILL.md
- Branch
- main
- Updated
- 2026-09-19
Related skills
- bdi-mental-states — This skill should be used when modeling agent mental states with BDI concepts: beliefs, desires, intentions, RDF-to-belief transformations.
- project-development — This skill should be used for project-level decisions about LLM-powered systems: whether an LLM is the right primitive for the task at hand.
- tool-design — This skill should be used for the tool-interface layer of an agent system specifically: writing tool descriptions agents can route on.
- legacy-modernizer — Designs incremental migration strategies, identifies service boundaries, produces dependency maps and migration roadmaps.
- advanced-evaluation — This skill should be used for advanced LLM evaluation: LLM-as-judge systems, direct scoring, pairwise comparison, rubric calibration.
- context-compression — This skill should be used when long-running agent sessions need context compression, structured summarization, compaction, token-per-task optimization.