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/modeled-information-format/mnemonic/memory-filesgit clone --depth 1 https://github.com/modeled-information-format/mnemonicWhat 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 | $0.00412 | $0.00412 |
| Opus 5 | $0.00206 | $0.00206 |
| Sonnet 5 | $0.00082 | $0.00082 |
| Haiku 4.5 | $0.00041 | $0.00041 |
Grade A, and why
mnemonic memory-files.instructions.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 2d 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.
What it actually says
Memory File Guidelines (MIF Level 3)
Required Frontmatter Fields
---
id: <uuid> # Unique identifier (uuidgen format)
type: semantic|episodic|procedural
namespace: {category}/user # e.g., decisions/user, learnings/user
created: <ISO8601> # Creation timestamp
modified: <ISO8601> # Last modification timestamp
title: "Title" # Human-readable title
tags: [tag1, tag2] # Categorization tags
---
Optional Temporal Fields
temporal:
valid_from: <ISO8601> # When content became valid
valid_until: <ISO8601> # When content expires (optional)
recorded_at: <ISO8601> # When it was recorded
decay:
model: exponential # Decay model type
half_life: P7D # ISO 8601 duration
strength: 0.85 # Current strength 0.0-1.0
Optional Provenance Fields
provenance:
source_type: user_explicit|inferred|conversation
agent: model-identifier # e.g., claude-opus-4
confidence: 0.9 # Confidence score 0.0-1.0
Memory Types
- semantic: Facts, decisions, specifications - things that ARE true
- episodic: Events, debug sessions, incidents - things that HAPPENED
- procedural: Workflows, patterns, how-tos - things you DO
Content Structure
After frontmatter, use standard Markdown:
- H1 heading matching the title
- Clear sections for context, details, implications
- Code blocks with language specification
- Links to related memories or external resources
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.
- 2d ago First seen · 56 lines · 412 tokens per session scan A d8526eb08b9b
mnemonic memory-files.instructions.md is an instructions file published in the GitHub repository modeled-information-format/mnemonic (22 stars, last pushed 1mo ago), licensed MIT. It adds 412 tokens to every session, about $0.0021 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
kipi-system AGENTS.md
AGENTS.md instructions for assafkip/kipi-system, covering q entrepreneur os, project structure, conventions, commands and build and test.
kipi-system CLAUDE.md
Claude Code instructions for assafkip/kipi-system, covering q entrepreneur os, project structure, conventions, commands and build and test.
personalos-boilerplate AGENTS.md
Instructions for vincentmumme/personalos-boilerplate, covering personalos boilerplate, einstieg mit einem nutzer, verbindliche grenzen, maintainer-arbeit and sprache.
brain-starter CLAUDE.md
Instructions for michaeljauk/brain-starter, covering brain repo - claude instructions, purpose, not set up yet?, file placement rules and directory guide.
markdown-agent-vault-ru AGENTS.md
Instructions for dzhokhov/markdown-agent-vault-ru, covering agents.md — правила агента для файлового хранилища, главный принцип, карта папок, загрузка контекста and недосказанные запросы.
context-canopy CLAUDE.md
Instructions for YusenZhang0601/context-canopy, a project described as: Portable, local-first AI agent memory and personal context—switch agents without losing identity, goals, rules, or evidence.