ai-memory-developer
Skill8 starsChanged 6 months agoMITView on GitHub
Guides design and implementation of AI agent and application memory—short-term conversation state, long-term user/tenant memory, episodic vs semantic stores, consolidation, retrieval, forgetting, privacy retention, and eval of memory quality. Use when building persistent memory for copilots, designing memory APIs, choosing vector or graph stores, implementing memory write/read policies, debugging wrong or stale memories, or tuning what the model should remember across sessions—not for general RAG document search (ai-engineer), context window packing and token budgets (ai-context-engineer), AI team operations cadence (ai-lead-ops), or org-wide token cost improvement roadmaps (ai-token-improvement-plan-engineer).
1 other project uses a version of this.
Adopted by
Most recently taken up by
- rade-nugroho/Agentic-Enteprises-Skillword for word, first seen May 20, 2026
Other versions1 project uses a version of this.
Word for word · 1
- rade-nugroho/Agentic-Enteprises-Skill★ 9100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository197 other files it gives its agents.
- ab-testing-engineerSkill
- actuarial-analystSkill
- actuarial-consultingSkill
- actuarySkill
- advanced-long-term-actuarial-mathematicsSkill
- advanced-persistent-threatSkill
- advanced-short-term-actuarial-mathematicsSkill
- agentic-ai-developerSkill
- ai-adversarial-robustness-engineerSkill
- ai-context-engineerSkill
- ai-engineerSkill
- ai-lead-opsSkill
- ai-redteamSkill
- ai-researcherSkill
- ai-risk-governanceSkill
- ai-skill-managerSkill
- ai-token-improvement-plan-engineerSkill
- aml-cftSkill
- aml-complianceSkill
- analytics-data-engineering-manager-productSkill
- analytics-data-engineerSkill
- anti-false-positive-decision-makingSkill
- applied-ai-architect-commercial-enterpriseSkill
- appointed-chief-actuarySkill
- asset-liability-managementSkill
- associate-actuarySkill
- assumption-settingSkill
- auditorSkill
- bcm-disaster-recovery-specialistSkill
- bi-analystSkill
- build-validatorSkill
- business-analystSkill
- business-consultantSkill
- business-model-researcherSkill
- certified-information-systems-security-professionalSkill
- chief-information-security-officerSkill
- ci-cd-engineerSkill
- cicd-engineerSkill
- cisco-certified-network-professionalSkill
- classified-cyber-security-senior-managerSkill
- classified-software-devsecops-engineerSkill
- cloud-architectSkill
- cloud-compliance-specialistSkill
- cloud-economistSkill
- cloud-engineerSkill
- cloud-security-engineerSkill
- cloud-system-administratorSkill
- cluster-deployment-engineerSkill
- code-securitySkill
- commercial-counselSkill
- communication-leadSkill
- community-executive-escalations-program-managerSkill
- compliance-engineerSkill
- compliance-specialistSkill
- compute-accounting-managerSkill
- control-software-developerSkill
- corporate-counselSkill
- country-managerSkill
- cross-department-translationSkill
- cryptographer-specialistSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 101
Run nightly consolidation: merge episodic notes into semantic summaries.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
AI Memory Developer
When to Use
- Building persistent memory for copilots, agents, or conversational AI
- Designing memory APIs (read/write/consolidate/forget)
- Choosing between vector stores, graph databases, or structured DBs for memory
- Implementing memory write/read policies and ACLs
- Debugging wrong, stale, or hallucinated memories
- Tuning what the model should remember across sessions (episodic vs semantic)
- Planning GDPR deletion paths and privacy retention for stored memories
- Evaluating memory quality (recall, precision, isolation)
When NOT to Use
- General RAG document search or indexing pipelines →
ai-engineer - Context window packing, token budgets, or compression →
ai-context-engineer - AI team operations, release governance, or SLOs →
ai-lead-ops - Org-wide token cost improvement roadmaps →
ai-token-improvement-plan-engineer
Related skills
| Need | Skill |
| End-to-end LLM app, RAG, agents | ai-engineer |
| Context assembly and compression | ai-context-engineer |
| Prompt and tool message design | prompt-engineer |
| PII retention and governance | ai-risk-governance |
| Production monitoring and incidents | ai-lead-ops |
| Token cost program and phased savings plan | ai-token-improvement-plan-engineer |
Core Workflows
1. Memory model design
Classify memory types:
| Type | Lifetime | Examples | Store |
| Working | Single turn / tool loop | Tool results, scratchpad | In-context only |
| Session | Chat session | Current task state | Redis / thread store |
| User long-term | Cross-session | Preferences, facts user stated | Vector + structured DB |
| Organizational | Shared | Docs, policies | RAG index (see ai-engineer) |
Design decisions:
- What may be written automatically vs requires user confirmation?
- Per-tenant isolation and ACL on every read/write
- TTL and deletion (GDPR erase path)
- Conflict resolution when new fact contradicts old
See references/memory_architecture.md for patterns and anti-patterns.
2. Write path (ingestion to memory)
observe → extract candidates → score importance → dedupe → persist → indexChecklist:
- Extract only durable facts, not transient chit-chat
- Attach provenance (message ID, timestamp, source)
- Dedupe with embedding similarity + entity linking
- Never store secrets, raw payment data, or full medical records unless required and approved
See references/write_consolidation.md for extraction prompts and consolidation jobs.
3. Read path (retrieval for generation)
- Build query from current user message + session summary
- Retrieve top-k memories with metadata filters (
user_id,tenant_id) - Rerank; drop below relevance threshold
- Inject into context in structured block (see
ai-context-engineer) - Cite memory IDs in logs for debugging
See references/read_retrieval.md for ranking and injection formats.
4. Forgetting and maintenance
| Trigger | Action |
| User delete request | Hard delete all user memories |
| TTL expired | Archive or purge |
| Low usefulness score | Decay or summarize away |
| Contradiction | Supersede old record; keep audit trail |
Run nightly consolidation: merge episodic notes into semantic summaries.
See references/write_consolidation.md for consolidation algorithms.
5. Evaluation
| Test | Pass criteria |
| Write accuracy | Gold facts appear in store after session |
| Recall | Question answerable from prior session |
| Precision | Irrelevant memories not retrieved |
| Isolation | Tenant A never sees tenant B |
| Forgetting | Deleted user has zero retrievable memories |
See references/memory_eval.md for datasets and regression harness.
When to load references
- Architecture and stores →
references/memory_architecture.md - Write and consolidation →
references/write_consolidation.md - Read and ranking →
references/read_retrieval.md - Evaluation →
references/memory_eval.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.