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/moai-team-llc/agenticmind/agents-mdgit clone --depth 1 https://github.com/Moai-Team-LLC/AgenticMindWrote 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/moai-team-llc/agenticmind/agents-md)<a href="https://agentmods.dev/instructions/moai-team-llc/agenticmind/agents-md"><img src="https://agentmods.dev/badge/instructions/moai-team-llc/agenticmind/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 | $0.01118 | $0.01118 |
| Opus 5 | $0.00559 | $0.00559 |
| Sonnet 5 | $0.00224 | $0.00224 |
| Haiku 4.5 | $0.00112 | $0.00112 |
Grade A, and why
AgenticMind 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 4d 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 — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidance for AI agents and coding assistants working in this repository.
What this is
AgenticMind is an auditable, self-improving knowledge & memory substrate for AI agents, served over the Model Context Protocol (MCP). It is not an agent and has no agent loop — the consuming product owns that. AgenticMind owns one concern: answers an agent can trust (citation-enforced + a replayable why-trace), a memory that compounds (judge-gated), exposed as MCP tools.
It implements a standard — read it first
AgenticMind is the reference implementation of the Agentic Product Standard. If you are designing or building an agentic product (architecture, harness, memory, evals, tools/MCP, durable execution), read the canon and use its skills:
➡️ https://github.com/Moai-Team-LLC/agentic-product-standard
When a product needs the memory / knowledge layer (the Standard's Layer 4), AgenticMind is that layer — point the agent's MCP client at it instead of reinventing retrieval, grounding, and a self-improving corpus.
Commands
bun run check # format + lint + tsc + tests (the full gate)
bun test # unit tests
bun run tsc # typecheck
bun run build-eval # regenerate the eval suite (self-validating)
bun run seed-eval # ingest the eval corpus into a DB
bun run eval # run the eval suite (needs DB + a chat key)
bun run reembed # recompute embeddings after a dimension change
Lint (oxlint) needs Node ≥22.18 (see .nvmrc); the rest run under Bun.
Conventions (please honor)
- Conventional Commits, enforced by commitlint + a husky
commit-msghook and a PR CI check. Header ≤72 chars. SeeCONTRIBUTING.md. - The MCP tool contract is versioned and frozen (
MCP_CONTRACT_VERSION,CONTRACT.md). A snapshot test guards it — any change to a tool's name/shape must bump the version and update the test in the same change. - Every new failure mode becomes a permanent eval case. Don't add behavior without a test; don't add a tool without a contract entry.
- No secrets, PII, or domain-private data in code, tests, or fixtures.
- Functional style,
neverthrowResult types, strict TypeScript.
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.
- 4d ago First seen · 88 lines · 1,118 tokens per session scan A db01f7e84739
AgenticMind AGENTS.md is an instructions file published in the GitHub repository Moai-Team-LLC/AgenticMind (4 stars, last pushed 27d ago), licensed Apache-2.0. It adds 1,118 tokens to every session, about $0.0056 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 2026-08-31.
Other instructions, from other repositories
llm-wiki-agent CLAUDE.md
Claude Code instructions for SamurAIGPT/llm-wiki-agent, covering llm wiki agent — schema & workflow instructions, slash commands (claude code), directory layout, page format and ingest workflow.
hippocampus CLAUDE.md
Instructions for sturlese/hippocampus, covering layout, frontmatter schema, conventions, answering questions from the vault (read order) and contradictions.
pgmnemo AGENTS.md
AGENTS.md instructions for pgmnemo/pgmnemo, covering pgmnemo — agent integration guide, 1. what pgmnemo is — and the problem it solves, the problem, architectural consequences and 2. when to adopt / when not to.
memoriki CLAUDE.md
Instructions for AyanbekDos/memoriki, covering llm wiki - schema, directory structure, rules, operations and ingest.
LLM-Wiki AGENTS.md
Instructions for VectorPeak/LLM-Wiki, covering obsidian second brain - codex cli operating manual, how to operate, command routing tables (grouped by category), wiki layer - simplified knowledge operations and vault - daily writing, capture, find.
llm-knowledge-bases CLAUDE.md
Instructions for rvk7895/llm-knowledge-bases, covering claude.md -- project instructions, project identity, configuration, skills and formatting rules (obsidian-compatible).