Skills conversation-memory
📦

conversation-memory

Content revision r2 High Risk ⚙️ External commands

Build Persistent Memory for LLM Conversations

LLM applications lose useful context across turns and sessions. This skill provides patterns for tiered storage, entity facts, retrieval, consolidation, and user isolation.

Supports: Claude Codex Code(CC)
⚠️ 38 Poor

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request
Review the Skillstore skill "conversation-memory" from https://skillstore.io/skills/sickn33-conversation-memory.md and its manifest at https://skillstore.io/api/skills/sickn33-conversation-memory/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/sickn33-conversation-memory.md Signed manifest GET /api/skills/sickn33-conversation-memory/manifest Signed lockfile GET /api/skills/sickn33-conversation-memory/lockfile

Test it

Using "conversation-memory". Design memory for a customer support assistant.

Expected outcome:

  • Buffer: Keep the latest conversation turns for immediate continuity.
  • Session memory: Retain active case details until resolution.
  • Long-term memory: Store consented preferences and verified account facts.
  • Entity memory: Track products, cases, and contacts with provenance.

Using "conversation-memory". How should old memories be managed?

Expected outcome:

Score memories by importance, consolidate valuable items, expire low-value records, enforce storage limits, and preserve user deletion requirements.

Using "conversation-memory". Review retrieval safety for a multi-user assistant.

Expected outcome:

Require authenticated user scope on every key and query. Verify ownership before updates, exports, or deletion, and test for cross-user leakage.

Security Audit

High Risk
v5 • 7/23/2026 Open versioned report

All 14 static findings are false positives caused by JavaScript template literals, Markdown backticks, and ordinary exception handling. However, examples directly interpolate untrusted messages and recalled memories into prompts, creating a persistent prompt-injection risk.

1
Files scanned
500
Lines analyzed
0
Review items
0
False positives ignored

Confirmed security concerns (1)

High
SKILL.md:145-152SKILL.md:214-231
Persistent Prompt Injection Through Untrusted Memory
Examples interpolate messages and recalled memories directly into prompts without trust labeling or instruction filtering. Attackers could store directives that influence later model behavior.
The examples visibly interpolate message, entity, and memory content into LLM prompts without a trust boundary. Persistence makes injected instructions reusable across sessions.

Risk Factors

⚙️ External commands (13)
SKILL.md:102 SKILL.md:109 SKILL.md:145-152 SKILL.md:214-218 SKILL.md:219 SKILL.md:220-222 SKILL.md:223 SKILL.md:224-231 SKILL.md:343-348 SKILL.md:384 SKILL.md:478-483 SKILL.md:483-487 SKILL.md:487
Audited by: codex View Audit History →
Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report
Security Assessment

Copy report link

https://skillstore.io/skills/sickn33-conversation-memory/audits/5?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/sickn33-conversation-memory/security.svg)](https://skillstore.io/skills/sickn33-conversation-memory?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/sickn33-conversation-memory?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/sickn33-conversation-memory/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/sickn33-conversation-memory.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
Academic citations (APA · BibTeX · CFF)

APA citation

sickn33. (2026). conversation-memory security audit report (audit version 5) [Author version unspecified]. Skillstore. https://skillstore.io/skills/sickn33-conversation-memory/audits/5

BibTeX citation

@techreport{sickn33-sickn33-conversation-memory-2026, author = {sickn33}, title = {conversation-memory security audit report (audit version 5)}, institution = {Skillstore}, year = {2026}, number = {5}, url = {https://skillstore.io/skills/sickn33-conversation-memory/audits/5}, note = {Author version unspecified} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "conversation-memory security audit report (audit version 5)" version: "unspecified" type: report authors: - name: "sickn33" date-released: "2026-07-23" url: "https://skillstore.io/skills/sickn33-conversation-memory/audits/5" identifiers: - type: other value: "skillstore:sickn33-conversation-memory:audit:5" description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium
55
Architecture
85
Maintainability
87
Content
69
Community
91
Spec Compliance

What You Can Build

Add Memory to a Chatbot

Plan memory tiers that preserve recent context, durable preferences, and important facts across sessions.

Design a Scalable Memory Service

Define retrieval, consolidation, storage limits, and cleanup policies for a production conversation platform.

Protect Tenant Memory

Review namespacing, access filters, ownership checks, exports, and deletion flows for privacy risks.

Try These Prompts

Choose Basic Memory Tiers 
Design a basic memory system for my [application type]. Separate recent conversation, session memory, durable facts, and entity details.
Add Retrieval and Consolidation 
Review this memory design: [design]. Add relevance retrieval, importance scoring, consolidation schedules, cleanup limits, and clear retention rules.
Design Tenant Isolation 
Create a tenant-isolated memory architecture for [platform]. Cover authenticated identifiers, scoped keys, filtered queries, ownership checks, exports, and deletion.
Threat Model Persistent Memory 
Threat model this LLM memory workflow: [workflow]. Address prompt injection, memory poisoning, personal data, provenance, authorization, retention, and evaluation tests.

Best Practices

  • Store only useful, consented information with provenance and retention metadata.
  • Scope every memory operation to an authenticated user or tenant.
  • Treat recalled content as untrusted data and evaluate retrieval quality regularly.

Avoid

  • Do not store every message indefinitely without importance filtering or cleanup.
  • Do not share storage namespaces or run retrieval without mandatory user filters.
  • Do not place recalled text into privileged prompts without prompt-injection controls.

Frequently Asked Questions

What memory types does this skill cover?
It covers conversation buffers, short-term memory, long-term memory, and structured facts about entities.
Does this skill provide a complete implementation?
No. It provides design patterns and illustrative TypeScript-like examples that require adaptation, testing, and infrastructure.
Which storage system should I use?
Choose storage based on latency, durability, search, privacy, and scale requirements. Redis and dedicated memory layers are mentioned as options.
How does the skill prevent cross-user data leakage?
It recommends authenticated user identifiers, namespaced keys, mandatory query filters, ownership checks, and user-specific export and deletion.
How should memory growth be controlled?
Use importance thresholds, capacity limits, scheduled consolidation, expiration, and deletion policies.
Does the guidance fully prevent prompt injection?
No. Add trust labels, instruction filtering, provenance checks, structured extraction validation, and adversarial tests before production use.

Developer Details

Author

sickn33

License

MIT

Skillstore revision

r2

Version notice

The author did not declare a version.

Repository

https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/conversation-memory

Ref

f9e2c34b4f19c7f3e6b0a1e93227b5f77cc12526

Maintenance freshness

7/26/2026

Usage

11 downloads · 149 views

File structure

📄 SKILL.md