Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add d-padmanabhan/agent-engineering-handbook --skill memory-architecturegit clone --depth 1 https://github.com/d-padmanabhan/agent-engineering-handbookWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/d-padmanabhan/agent-engineering-handbook/memory-architecture)<a href="https://agentmods.dev/skills/d-padmanabhan/agent-engineering-handbook/memory-architecture"><img src="https://agentmods.dev/badge/skills/d-padmanabhan/agent-engineering-handbook/memory-architecture/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/d-padmanabhan/agent-engineering-handbook/memory-architecture"><img src="https://agentmods.dev/badge/skills/d-padmanabhan/agent-engineering-handbook/memory-architecture.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00073 | $0.01834 |
| Opus 5 | $0.00036 | $0.00917 |
| Sonnet 5 | $0.00015 | $0.00367 |
| Haiku 4.5 | $0.00007 | $0.00183 |
Grade A, and why
memory-architecture scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 174 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Memory Architecture
Design persistent memory as a governed knowledge system, not as a larger prompt. Separate evidence from claims, choose storage from durability and trust requirements, filter before retrieval, cite what is used, and measure whether memory improves outcomes.
When to Use
Use this skill for:
- Agent or assistant memory that persists across sessions
- Durable project or team knowledge bases
- Retrieval pipelines that assemble context automatically
- Note deduplication, contradiction handling, supersession, or retention
- Security and privacy reviews of persistent AI memory
Do not use it for:
- One-off questions or ordinary documentation
- Current-task handoff files such as
tmp/active-context.md - Web research refreshes whose storage architecture is already defined
- Model-provider API syntax without a memory-system design decision
For temporary coding-agent context, use agent workflow context management. For bounded web refreshes, use web-research-kb-refresh. For trust-boundary reviews, compose with zero-trust.
Non-Negotiables
- Raw input is untrusted evidence, not ground truth. Preserve its provenance and scan it for prompt injection, sensitive data, and malformed content.
- Ephemeral context is not durable memory. A gitignored
tmp/directory is appropriate for replaceable session state, not the only copy of team knowledge. - Evidence and claims are separate records. A claim cites evidence; changing a claim does not rewrite the source.
- Trust never escalates automatically. Models and background jobs may propose claims or flag conflicts, but only an authorized human or deterministic policy may approve higher-trust states.
- Tenant and authorization filters run before semantic retrieval. Similarity does not grant access.
- Similarity is a candidate signal. It does not prove duplication, agreement, or contradiction.
- Supersession preserves lineage. Do not move or delete records in a way that breaks stable identifiers and citations.
- Every retrieved claim is attributable. Return stable claim IDs and source citations with generated context.
- Deletion is a designed operation. Remove the canonical record, derived chunks, embeddings, caches, and replicas according to policy.
- Quality is measured. Evaluate retrieval relevance, groundedness, freshness, conflict handling, and isolation before calling the system reliable.
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 174 lines · 73 tokens per session scan A 8bfd7f1e8ebc
memory-architecture is a skill published in the GitHub repository d-padmanabhan/agent-engineering-handbook (17 stars, last pushed today), licensed MIT. It adds 73 tokens to every session and 1,834 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
media-ingest
Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.
mem0-oss-to-platform
Plan and then execute a migration of a project from the mem0 open-source / self-hosted SDK (the local Memory class) to the mem0 Platform / hosted / managed SDK (the MemoryClient class). Use this whenever a developer wants to move, switch, or migrate their mem0 usage off OSS/self-hosted to the hosted API — e.g.…
Cortex
Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…
memory
Use when the user asks to remember, recall, forget, update, search, or inspect durable OpenSquilla memory, including profile facts in USER.md and long-term notes in MEMORY.md or memory//.md.
ha-data-stores
Map of Hope Agent's local data stores and safe read-only query workflow. Use when the user asks where Hope Agent stores data, wants to inspect sessions/messages/memory/logs/background jobs/knowledge indexes/settings, asks the model to query local app data, or debugging requires checking persisted state. Trigger…
establishing-project-context
Use when the user asks to establish shared project language, or project work exposes a conflicting, renamed, or deprecated domain term that needs active semantic modeling. Routine small tasks stay on the fast path.