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 rules/robotaitai/project-bedrock/shared-memorygit clone --depth 1 https://github.com/robotaitai/project-bedrockWhat 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 | $0.00960 | $0.00960 |
| Opus 5 | $0.00480 | $0.00480 |
| Sonnet 5 | $0.00192 | $0.00192 |
| Haiku 4.5 | $0.00096 | $0.00096 |
Grade A, and why
shared-memory 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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Shared Memory
This project stores persistent memory at agent-knowledge/Memory/MEMORY.md.
The repo-local ./agent-knowledge path is the working entrypoint and should point to the real dedicated knowledge folder.
If Cursor project memory is used, ~/.cursor/projects/<project-id>/memory/ should point to ./agent-knowledge/Memory/ so all agents read and write the same curated files.
When the symlink is missing or broken, create it:
ln -s <workspace>/agent-knowledge/Memory ~/.cursor/projects/<project-id>/memory
Tree structure
Memory is organized as a tree using the same-name branch convention. Agents load only the branches relevant to their current task.
agent-knowledge/
Memory/
MEMORY.md <- root curated-memory index -- always loaded, max ~200 lines
<topic>.md <- flat branch note (no subtopics)
<topic>/
<topic>.md <- branch entry note (same name as folder)
<subtopic>.md <- leaf note beside the entry
decisions/
decisions.md <- decision log
Evidence/
raw/
imports/
Reading the tree
Memory/MEMORY.mdis always loaded automatically. It is a concise index only -- no deep detail.- At session start, identify which branch(es) your task touches from the Branches section.
- Read the relevant branch entry notes before starting work.
- If the branch has subtopic files needed for your task, read those too.
- Do not read branches unrelated to your task -- keep context lean.
Writing to the tree
- Memory/MEMORY.md: short branch summaries with relative links. Keep it under 200 lines.
- Branch entry note (
<topic>.mdor<topic>/<topic>.md): stable facts for one focused domain. - Subtopic note (
<topic>/<subtopic>.md): created only when a branch note grows beyond ~150 lines. - When adding a subtopic, replace the inline detail in the entry note with a link.
- Each note must stay under ~150 lines. Split if bigger.
- Link related notes with relative markdown links (e.g.,
See [perception](perception/perception.md)). - One branch = one functional domain. Do NOT lump unrelated subsystems into one file.
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 · 102 lines · 960 tokens per session scan A 7d9cc9ef06d8
shared-memory is a cursor rule published in the GitHub repository robotaitai/project-bedrock (44 stars, last pushed 12d ago), licensed MIT. It adds 960 tokens to every session, about $0.0048 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 cursor rules, from other repositories
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
typescript
Changes to these high-fan-out internals can affect every message, delta, element, or rerun. Keep work in them minimal, and benchmark changes with representative stress-test apps.
coolify-ai-docs
Master reference to all Coolify AI documentation in .ai/ directory.
python_lib
Tips and guidelines specific to the development of the Streamlit Python library, not applicable to scripts and e2e tests.
specs
This directory contains product and tech specs for Streamlit features.