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/marerem/longmem/claude-mdgit clone --depth 1 https://github.com/marerem/longmemWhat 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 | $0.02091 | $0.02091 |
| Opus 5 | $0.01045 | $0.01045 |
| Sonnet 5 | $0.00418 | $0.00418 |
| Haiku 4.5 | $0.00209 | $0.00209 |
Grade A, and why
longmem 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 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.
How it starts
The opening of the file, as written. The whole thing — 198 lines — stays where its author put it; the contents beside it link to each section on GitHub.
longmem
MCP plugin that stores problem/solution pairs in a local vector DB (LanceDB) and suggests cached solutions across projects using embedding similarity search.
Auto-memory workflow — follow every time
Step 1 — search before solving
Before reasoning about any bug, error, or implementation question, call:
search_similar(
problem="<clear description of what is broken or needed>",
category="<best matching category>",
tags=["<library>", "<tool>", "<error type>"],
language="<language if relevant>"
)
- Similarity ≥ 85%: present the cached solution first. Check
edge_cases— if any apply, say so upfront. - Similarity < 85%: solve normally, then go to Step 2.
Step 2 — save after confirmed solve
Wait until the conversation reaches a natural end (user confirms it works, no more follow-ups). Then synthesize and call:
confirm_solution(
solution="<synthesized, reusable answer — see format below>",
project="longmem"
)
Problem description, category, tags, and language are filled in automatically from the Step 1 call.
How to write solution for maximum reuse:
- State the general pattern or principle first
- Then give the specific detail from this session as an example
- Bad: "Port 4181 is used by the auth proxy for Paperless"
- Good: "Port 418x is typically an OAuth2 auth proxy frontend. Default is 4180. Use 4181+ when 4180 is already taken by another auth proxy in the same stack. Example: Sifonia on 4180, Paperless-NGX auth proxy on 4181."
If you did not call search_similar first (skipped Step 1), fall back to:
save_solution(
problem="<clear, reusable description — not tied to this repo>",
solution="<synthesized answer as above>",
category="<category>",
project="longmem",
tags=["<library>", "<tool>", "<error type>"],
language="<language>"
)
Step 3 — fix or enrich after saving
If entry_id is in context (same conversation as the save):
- Correction (wrong name/term):
correct_solution(entry_id=..., find=..., replace=...) - New facts/context:
enrich_solution(entry_id=..., context=...) - Failure/exception:
add_edge_case(entry_id=..., edge_case=...)
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 · 198 lines · 2,091 tokens per session scan A 18dddc94372d
longmem CLAUDE.md is an instructions file published in the GitHub repository marerem/longmem (1 stars, last pushed 4mo ago), licensed MIT. It adds 2,091 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
vscode buildNext.instructions.md
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).
spec-kit AGENTS.md
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.
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.
langchain AGENTS.md
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.
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).
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.