memory-design skillA
memory-design is agent-read markdown (skill) from nahid-sparktales/agent-dispatcher: Decide what an agent should remember, which layer holds it, who it is scoped to, and how a stale or contradicted memory is detected and retired. Use when an agent forgets something across sessions, when a memory or persistent-context feature is being designed, or when stored memories have grown noisy, wrong, or are leaking between users. Not for retrieval over a document corpus, not for prompt or context-window packing, and not for conversation transcript storage..
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file says
# Memory design Memory is the only part of an agent that keeps being wrong after you stop looking. A retrieval bug shows up in the next answer; a stale memory sits there being quietly confident for months. Design the retirement path before the write path. ## When this fires An agent needs a fact to survive past the current context — across turns, tasks or sessions — or an existing memory store has grown noisy, contradictory, or is surfacing one subject's data to another. It does not fire for retrieval over a document corpus (that is retrieval-rag), and not for keeping a transcript: a transcript is a log, not memory. ## Procedure 1. **Name the repeat failure first.** What did the agent forget, how often, and what did it cost — a re-asked question, a redone decision, a wrong assumption. Memory added without a named repeat failure is speculative storage, and storage you cannot justify is storage you will not prune. 2. **Classify what is actually worth keeping**, because the class decides everything after it: - *stable facts and preferences* — this user's stack, conventions, how they want to be addressed …
Read the whole file at its exact version.
How to install
mdr add nahid-sparktales/agent-dispatcher/memory-design@git:20260919.a0d4f55mdr add nahid-sparktales/agent-dispatcher/memory-design@sha256:c4713e0e685cc728Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
[](https://markdownregistry.com/a/art_jw4uafrtwdwjcbeo)
1 badge views in 30 days
Versions
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (7424 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
nahid-sparktales/agent-dispatcher · 49 stars · license MIT · pushed 2026-09-23 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_jw4uafrtwdwjcbeo GET https://markdownregistry.com/api/v1/resolve?ref=nahid-sparktales/agent-dispatcher/memory-design GET https://markdownregistry.com/api/v1/blob/c4713e0e685cc7288f5283669b40e34403e55107fcd8180dfbfe56afe5d301b1
Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.