memory-load
Skill0 starsChanged 6 months agoMITView on GitHub
Loads relevant entries from memory/ into the conversation context. Use at the start of any non-trivial conversation, and any time the topic might overlap with what is stored in memory (personal facts, habits, goals, relationships, work, health, hobbies, etc.). Hybrid strategy — always read memory/index.md and memory/tags.md as the map first, then pull full entry bodies by tag match and by semantic match on titles. No artificial cap on how many entries can be loaded.
In this repository9 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- grill-meSkill
- listsSkill
- memory-retroSkill
- memory-writeSkill
- movie-wishlistSkill
- my-scheduleSkill
- new-daySkill
- tag-pickSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 4 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
When the user explicitly asks to recall — *"what did I say earlier about X?"*
- Distinctive: line 34
Use both signals together — tags for precise hits, titles for cases where the wording diverges.
- Very specific: line 38
Read the full .md files of the entries that passed the filter.
- Distinctive: line 45
When citing a specific fact, refer to it briefly (*"you said earlier that …"*) without quoting the full formulation.
- Distinctive: line 51
Do not write or modify memory files.
- Very specific: line 52
If something needs to be recorded — that is memory-retro's job.
≈ “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
memory-load
Pulls relevant memory entries into the conversation context.
When to invoke
- At the start of a session or whenever a new initiative is opened — required.
- Any time the conversation touches a topic that could plausibly be in memory (personal, habits, goals, relationships, work, health, hobbies, family, etc.).
- When the user explicitly asks to recall — "what did I say earlier about X?"
Err on the side of invoking. A missed context is worse than a redundant read.
Algorithm
Step 1. Read the map
Always read:
memory/index.md— the catalog of all entries with titles and tags.memory/tags.md— the tag dictionary (needed to interpret tags).
These files are small and cheap. Read them in full.
Step 2. Decide relevance
From the user's last message and the conversation context, extract:
- Candidate tags — which existing entries in
tags.mdmatch the topic. - Semantic hooks — entry titles in
index.mdthat are thematically close even when the tags themselves do not line up (the user may use different words).
Use both signals together — tags for precise hits, titles for cases where the wording diverges.
Step 3. Load the bodies
Read the full .md files of the entries that passed the filter.
Default filter: status: active only. Entries with status: deprecated are pulled in only when the user is explicitly asking about the past ("what did I used to think about X", "go back to the old idea about Y").
Step 4. Use in the response
- Lean on the loaded knowledge implicitly. Do not recite entries verbatim unless the user asks.
- When citing a specific fact, refer to it briefly ("you said earlier that …") without quoting the full formulation.
- Respect
confidence:lowandmediumare in-the-moment hypotheses, not doctrine.lowentries should be revisited and confirmed at the next natural opportunity. - Respect
type:constraint— a hard limit on recommendations;goal— a direction to nudge toward;belief— an opinion that can be discussed;preference— color the style and content of suggestions;fact— given;event— historical anchor.
Constraints
- Do not write or modify memory files. This skill is read-only.
- If something needs to be recorded — that is
memory-retro's job.
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.