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/elvirafa/agentic-memory/agents-mdgit clone --depth 1 https://github.com/elViRafa/agentic-memoryWhat 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 | $0.02094 | $0.02094 |
| Opus 5 | $0.01047 | $0.01047 |
| Sonnet 5 | $0.00419 | $0.00419 |
| Haiku 4.5 | $0.00209 | $0.00209 |
Grade A, and why
agentic-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 yesterday.
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 — 172 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Instructions — Memory Fabric
This file is read automatically by Claude Code, Gemini CLI, Codex, Antigravity, and other MCP-aware AI agents.
GitHub Copilot reads .github/copilot-instructions.md instead.
Memory Fabric — Semantic Store Agent Instructions
🚨 CRITICAL RULES - READ FIRST 🚨
- NEVER use the native VS Code Copilot
memorytool. You MUST ONLY use thememory-fabricMCP tools (likewrite_memory_store_tool). The nativememorytool writes to VS Code workspace storage, bypassing this project's memory system. - NEVER use raw file system tools (like
create_file,write_to_file,bash, etc.) to read or write files inside the.ai-memory/directory. Doing so bypasses secret scanning, token budgeting, and the Dreaming system. - MANDATORY STARTUP: You MUST call
read_combined_context_tool(cwd="<absolute project root path>")before doing anything else at the start of a session. No exceptions.MCP Resources alternative: If your client supports MCP Resources and has auto-fetched
memory-fabric://context/<encoded-cwd>, that context is already in your system prompt — skip the tool call. - NEVER call
dream_toolas a substitute for saving new knowledge. Before triggering any Dream tool, you MUST first callwrite_memory_store_toolto persist specific, isolated memories from the current session (e.g., bugs fixed, features built, architecture decisions). Dreaming consolidates existing memory — it does NOT capture new knowledge. - MANDATORY SESSION END: Before your final response in a session, you MUST call
write_session_journal_toolto log what was accomplished. Skip ONLY for trivial Q&A sessions with no code changes, decisions, or debugging.
1. Active Retrieval Workflow
- Search: Use
keyword_search_tool(cwd, query)to find specific documented topics. - Deep Dive: Use
read_memory_store_tool(cwd, store_path)orread_section(cwd, section)for detailed content.
2. Store Writes & Rules
Use write_memory_store_tool to register standalone memories.
store_pathRules: Must be lowercase, alphanumeric segments separated by slashes. No spaces, capitals, or.mdextension (e.g.,architecture/decisions/jwt-auth). Max 5 levels of nesting.- Parameters:
cwd,store_path,content,title(optional),tags(optional),priority(high/medium/low),mode(replace/append).
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.
- yesterday First seen · 172 lines · 2,094 tokens per session scan A 8ba9ba039ebc
agentic-memory AGENTS.md is an instructions file published in the GitHub repository elViRafa/agentic-memory (2 stars, last pushed 14d ago), licensed MIT. It adds 2,094 tokens to every session, about $0.0105 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
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
buildNext
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
Instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
langchain AGENTS.md
Instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.