agent-memory-vault
Skill1 starsChanged 23 days agoMITView on GitHub
Structured, file-backed memory for autonomous agents that must recall decisions and preferences across runs. Use when a workflow needs durable facts that outlive a single session.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- wuzishu3-web/agent-memory-vaulttheir own version, first seen Jun 14, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- wuzishu3-web/agent-memory-vault★ 0same name
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 repository46 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .codex/AGENTS.mdAGENTS.md
- .gemini/AGENTS.mdAGENTS.md
- .cursor/rules/agent-foundry-orchestrator.mdcCursor rule
- .cursor/rules/agent-foundry-skills.mdcCursor rule
- web/llms.txtllms.txt
- anti-slopSkill
- api-designSkill
- automation-pickSkill
- bottleneck-gatingSkill
- budget-guardSkill
- cache-identitySkill
- constraint-then-solveSkill
- context-optimizationSkill
- cron-troubleshootSkill
- e2e-test-strategySkill
- engram-routingSkill
- execution-observabilitySkill
- fallback-model-routingSkill
- feedback-loopSkill
- judge-integritySkill
- knowledge-extractSkill
- landscape-firstSkill
- measure-firstSkill
- plan-before-codeSkill
- plan-then-actSkill
- prompt-disciplineSkill
- pushback-when-wrongSkill
- quality-protocolSkill
- read-before-buildSkill
- re-verify-findingsSkill
- session-closeoutSkill
- session-distillSkill
- show-your-workSkill
- structured-output-contractSkill
- verify-firstSkill
- workflow-decomposeSkill
- agent-team-standardsSkill
- autonomous-briefingSkill
- book-to-skillSkill
- chrome-devtools-mcp-bridgeSkill
- design-languageSkill
- funnel-pr-guardSkill
- persistent-memorySkill
- sql-migration-trioSkill
- token-compressionSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 32
Run from the repository root:
- Distinctive: line 57
Run one writer at a time, or wrap calls in your scheduler's lock.
- Very specific: line 73
If Python is unavailable, keep one topic.jsonl file and append one JSON object per line with a key and value.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
Agent Memory Vault
A small JSONL-backed fact store for agents. It records keyed values per topic and returns recent context at the start of a later run. It is not a transcript log, a vector index, or a multi-writer database.
Procedure
What ships
The script is skills/optional/agent-memory-vault/scripts/memory.py. It uses
only the Python standard library (Python 3.10+).
Storage defaults to ~/.agent-memory/; set AGENT_MEMORY_HOME to relocate it.
Each topic is an append-only .jsonl file plus an index.json summary.
Commands
Run from the repository root:
python skills/optional/agent-memory-vault/scripts/memory.py put \
--topic=preferences --key=code-style --value="2-space indent"
python skills/optional/agent-memory-vault/scripts/memory.py get \
--topic=preferences --key=code-style
python skills/optional/agent-memory-vault/scripts/memory.py list --topic=preferences
python skills/optional/agent-memory-vault/scripts/memory.py context --last=10Topics
| Topic | What goes here |
| preferences | User preferences (style, tools, languages) |
| decisions | Architecture or design choices with rationale |
| project-state | Work in progress, blockers, recent changes |
| errors | Errors encountered and fixes |
| glossary | Domain-specific terms |
Concurrency limit
The script appends entries and then rewrites index.json without a file lock.
Two writers at the same time can interleave lines or lose an index update.
Run one writer at a time, or wrap calls in your scheduler's lock. Treat this as
a single-writer store, not a shared database.
Cron example
# Daily at 09:00: load preferences, then append findings.
0 9 * * * cd "$HOME/agent" && \
<your-agent-command> "Load memory topic preferences; append today's findings to topic briefing."Replace <your-agent-command> with your harness command. No scheduler or
agent loop ships with this skill.
Manual fallback
If Python is unavailable, keep one topic.jsonl file and append one JSON object
per line with a key and value. The context command will not parse it, so
read the last N lines directly and state that the index is stale.
Anti-patterns
- Storing secrets, tokens, or credentials in the vault.
- Writing free-form notes without a topic key; they become unsearchable.
- Running multiple cron jobs against the same topic without an external lock.
- Treating the vault as a transcript or as project documentation.
Verification Checklist
putfollowed bygetreturns the same value.list --topic=<topic>shows the expected entry count.context --last=10includes the new entry and no secret values.- A second writer was not running during the test.
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.

