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 agentmods add instructions/zycaskevin/vault-agent-memory/agents-mdgit clone --depth 1 https://github.com/zycaskevin/Vault-Agent-MemoryWrote 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/instructions/zycaskevin/vault-agent-memory/agents-md)<a href="https://agentmods.dev/instructions/zycaskevin/vault-agent-memory/agents-md"><img src="https://agentmods.dev/badge/instructions/zycaskevin/vault-agent-memory/agents-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.01618 | $0.01618 |
| Opus 5 | $0.00809 | $0.00809 |
| Sonnet 5 | $0.00324 | $0.00324 |
| Haiku 4.5 | $0.00162 | $0.00162 |
Grade A, and why
Vault-Agent-Memory AGENTS.md 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 5d 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 — 183 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Instructions For Vault Agent Memory
This repository is designed to be installed, configured, and modified by agent
runtimes as well as by humans. If you are an agent working in this repo, treat
this file and agent_manifest.json as the operational source of truth.
If you can run shell commands for the user, do not stop at telling them how to install Vault Agent Memory manually. Ask the required scope/feature questions, run the install/configuration steps yourself, and finish with a smoke test.
Common Install Architecture
Hermes Agent, Codex, OpenCode, Claude Code, OpenClaw, n8n, and other MCP-capable or shell-capable agents should use the same basic shape:
choose projectDir -> choose optional features -> install vault -> configure CLI or stdio MCP -> verify search/read/propose
Only add runtime-specific adapters when the host needs custom tool registration
or UI metadata. The stable cross-agent contract is projectDir, the vault
CLI, vault-mcp, and candidate-first memory writes.
For MCP-capable agents, prefer the smallest tool profile that fits the job:
vault-mcp --project-dir /path/to/project --tool-profile core
core exposes only vault_search, vault_read_range, vault_memory_propose,
and vault_stats. Use review, remote, maintenance, or full only when
those extra tools are needed.
Second Decision: Optional Features
After choosing database scope, ask which optional capabilities the user wants. Do not install heavyweight or cloud-connected extras silently.
Recommended prompts:
| Feature | Default | Ask when | Install command |
|---|---|---|---|
core |
yes | Always. Local SQLite, Markdown, keyword search. | python -m pip install vault-for-llm |
mcp |
yes for MCP-capable agents | The runtime supports local stdio MCP. | python -m pip install "vault-for-llm[mcp]" |
obsidian_import |
no | User has an existing Obsidian vault to import and search through Vault. | built into core CLI |
semantic |
no | User wants embedding-backed semantic/hybrid retrieval. | python -m pip install "vault-for-llm[semantic]" |
supabase |
no | User wants optional remote sync/read paths. | python -m pip install "vault-for-llm[supabase]" |
dev |
no | You are modifying the repo, running benchmarks, or validating a PR. | python -m pip install -e ".[dev]" |
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.
- 5d ago First seen · 183 lines · 1,618 tokens per session scan A a69840771cca
Vault-Agent-Memory AGENTS.md is an instructions file published in the GitHub repository zycaskevin/Vault-Agent-Memory (48 stars, last pushed 8d ago), licensed Apache-2.0. It adds 1,618 tokens to every session, about $0.0081 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 2026-08-30.
Other instructions, from other repositories
rekal AGENTS.md
Instructions for janbjorge/rekal, covering rekal: agent instructions, what this project is, architecture, key rules and re-expansion bar.
rekal CLAUDE.md
Instructions for janbjorge/rekal, a project described as: Long-term memory for LLMs. MCP server backed by hybrid search in a single SQLite file.
memobase AGENTS.md
Instructions for maximskorohod/memobase, covering agents.md — установка плагина memobase, что это, способ b — через git (человек ставит вручную, без чата), безопасность and справочные факты (для сверки при установке).
okforge-webui AGENTS.md
AGENTS.md instructions for okforge/okforge-webui, covering agents.md — okforge-webui repo map, map and conventions.
LexiconLocal CLAUDE.md
Claude Code instructions for alexeymohr/LexiconLocal, covering lexiconlocal, read first, this repository is public. the operator's data is not, tech stack (fixed — do not substitute) and hard rules.
grimoire CLAUDE.md
Instructions for vedantggwp/grimoire, covering claude.md, plugin structure, core rules, workspace convention and what not to do.