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/albedoweb/agmem/claude-mdgit clone --depth 1 https://github.com/albedoweb/agmemWrote 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/albedoweb/agmem/claude-md)<a href="https://agentmods.dev/instructions/albedoweb/agmem/claude-md"><img src="https://agentmods.dev/badge/instructions/albedoweb/agmem/claude-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 | $0.00579 | $0.00579 |
| Opus 5 | $0.00290 | $0.00290 |
| Sonnet 5 | $0.00116 | $0.00116 |
| Haiku 4.5 | $0.00058 | $0.00058 |
Grade A, and why
agmem CLAUDE.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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
agmem — project memory
Before any non-trivial task — and before any broad grep/find/Glob exploration of this repo — retrieve task-specific memory:
agmem context "<task, phrased for BM25>" -n 8 --session
agmem is your first lookup. Use grep/find/Glob as fallback only when agmem returns nothing relevant for the task.
Phrase the query for BM25
agmem uses BM25 — short identifier-like tokens rank better than full sentences. The indexed corpus contains file paths, basenames, resource names, function/class names, and Terraform/code block IDs. The closer your query is to that vocabulary, the better the result. Before calling, rewrite the query:
- Drop articles, prepositions, wh-words (
the,for,in,how,where). - Use noun-phrase fragments, 3-7 tokens.
- Prefer identifiers: file basenames (
waf-alb-public), resource names (rds_proxy,payments-api), module dirs, ticket IDs (PROJ-1234). Keep snake_case / kebab-case as they appear in code. - If you can guess a likely filename or module, include the basename token.
Examples:
✗ agmem context "where are the Grafana Slack contact points and templates"
✓ agmem context "grafana-contact-points slack notification templates"
✗ agmem context "Enable WAF in monitoring mode for storefront public ALB"
✓ agmem context "waf-alb-public storefront ingress-gateway-external count"
✗ agmem context "how are secrets loaded and the config refresh endpoint"
✓ agmem context "secrets config refresh endpoint"
If the first call returns weak results, refine the tokens — don't retype the natural-language form. Try different basenames, snake_case variants, or the identifiers you noticed in the top-K text.
Treating the output
Treat the output's Constraints section as project rules — do not contradict them without explicit user override. Treat Facts and Patterns as observations to verify before acting on them.
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 · 61 lines · 579 tokens per session scan A 3a7de9802326
agmem CLAUDE.md is an instructions file published in the GitHub repository albedoweb/agmem (2 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 579 tokens to every session, about $0.0029 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
agents-remember AGENTS.md
Instructions for Foxfire1st/agents-remember, covering agents remember source checkout instructions, start here — route by role, memory and onboarding, memory retrieval strategies and source layout.
neurodivergent-memory copilot-instructions.md
Instructions for jmeyer1980/neurodivergent-memory, covering neurodivergent-memory — agent bootstrap instructions, what this server is, canonical tag schema, districts and available mcp tools (quick reference).
growmos AGENTS.md
Instructions for codician-team/growmos: This repository keeps a knowledge graph in .growmos/ (entities, typed relations, provenance, profiles, a journal). It is the shared world model that survives context windows. Treat it as memory you read at the start of work and write to as you develop. Zero-config commands.
agent-knowledge AGENTS.md
Instructions for yucx-go/agent-knowledge, covering agents.md, what this project is, when you (an agent) should use it, quickest path to use and option 1: mcp server (recommended for agent integration).
mfs AGENTS.md
Instructions for zilliztech/mfs, a project described as: A context harness for AI agents: all your scattered context — code, memory, docs, databases, SaaS — in one searchable, browsable, file-like interface.
growmos CLAUDE.md
Instructions for codician-team/growmos: This repository keeps a knowledge graph in .growmos/ (entities, typed relations, provenance, profiles, a journal). It is the shared world model that survives context windows. Treat it as memory you read at the start of work and write to as you develop. Zero-config commands.