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/ozgurkarahan/ai-agent-memory/ingestgit clone --depth 1 https://github.com/ozgurkarahan/ai-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/ozgurkarahan/ai-agent-memory/ingest)<a href="https://agentmods.dev/instructions/ozgurkarahan/ai-agent-memory/ingest"><img src="https://agentmods.dev/badge/instructions/ozgurkarahan/ai-agent-memory/ingest.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 | $0.03614 | $0.03614 |
| Opus 5 | $0.01807 | $0.01807 |
| Sonnet 5 | $0.00723 | $0.00723 |
| Haiku 4.5 | $0.00361 | $0.00361 |
Grade A, and why
ai-agent-memory ingest.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 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 — 255 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ingest Source — LLM-Compiled Wiki Ingestion (Karpathy Pattern)
Two triggers:
- "ingest" — single-source ingest. Follow Phase 1 → Phase 7 as written.
- "mass ingest" / "bulk ingest" / "ingest all ... from ..." — multi-source batch. MUST insert Phase 0 (Inventory) and Phase 2b.5 (Consolidation Planning) before compiling anything. See [[ingest]] consolidation section for the full pattern and rationale.
Resolve the memory root
- If
memory/schema.mdexists in the current workspace, setWIKI_ROOTtomemory/. - Else if
schema.mdexists, setWIKI_ROOTto the current directory. - Else follow the memory-wiki path declared in the current project's
AGENT.md. - If no folder containing both
schema.mdandindex.mdresolves, report the missing path and stop.
All relative paths below are under WIKI_ROOT.
Phase 0: Inventory (mass ingest only)
Before pulling content from files or any connected source tool:
- List every source you intend to ingest (OneNote sections, files, URLs, query topics).
- Save the inventory to the session folder or as a temp list. Do NOT start pulling yet.
- If the source system can't be enumerated, explicitly state the limitation and the coverage estimate before proceeding.
- Ask the user to confirm scope if the inventory is >10 sources — consolidation decisions are cheaper to agree on upfront than to refactor later.
Phase 1: Gather Context
- Read
schema.md— understand wiki structure, categories, naming rules - Read
index.md— get the full list of existing articles with paths (needed for backlink resolution) - Read
glossary.md— check existing terms - Read the source content — the file, topic, or conversation findings to ingest
Phase 2: LLM Compilation (this is YOU — use your reasoning)
Compile the source into structured wiki knowledge. Ask yourself:
2a. What type of knowledge is this?
- Project knowledge →
wiki/projects/{slug}.md(create or update) - Domain/technology knowledge →
wiki/domains/{slug}.md(create or update) - Reusable pattern →
wiki/patterns/{slug}.md(create or update) - Debugging lesson / gotcha →
wiki/lessons/{slug}.md(create or update) - Skill / command / triggerable workflow →
wiki/skills/{slug}.md(create or update) - Agent / subagent / role-based executor →
wiki/agents/{slug}.md(create or update) - Tool knowledge →
wiki/tools/{slug}.md(create or update; only for products, CLIs, SDKs, APIs, services, utilities) - Platform/env knowledge →
agent-config/platform.mdoragent-config/knowledge/{domain}.md(update) - If a source does not fit the installed taxonomy, ask before extending
schema.md; do not invent a private or organization-specific category.
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 · 255 lines · 3,614 tokens per session scan A 0355e433b00a
ai-agent-memory ingest.instructions.md is an instructions file published in the GitHub repository ozgurkarahan/ai-agent-memory (8 stars, last pushed 1mo ago), licensed MIT. It adds 3,614 tokens to every session, about $0.0181 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
cognirepo CLAUDE.md
Claude Code instructions for ashlesh-t/cognirepo, covering claude.md, key rules, session start sequence (run in this order), behavioral confirmation rule and personas (cognirepo-402, cognirepo-403).
aeon CLAUDE.md
Claude Code instructions for aeonfun/aeon, covering aeon, how aeon works, strategy, voice and soul file hierarchy (read in this order).
wayland-core copilot-instructions.md
Copilot instructions for FerroxLabs/wayland-core, covering ijfw rules, output discipline, memory routing, context discipline and cross-audit.
mcp-structured-memory CLAUDE.md
Claude Code instructions for nmeierpolys/mcp-structured-memory, a project described as: Structured Memory MCP Server.
inkwell-memory CLAUDE.md
Instructions for veronchenko/inkwell-memory, covering claude.md — inkwellmemory, layout, multi-tenant mode (inkwellmultitenant=1), conventions and testing.
memory-mcp-1file AGENTS.md
Instructions for pomazanbohdan/memory-mcp-1file, covering ⛔ l0 invariants (never violate under any circumstance), detection heuristic, 🔀 phase transition routing (⛔ blocking), gate-0: phase identification & loading and 🧾 proof-of-load requirement (critical).