cursorrules

Rules for using the Firebolt Memory Layer, a service that stores and recalls conversation context. They define how sessions, working memory, and longer-term memories should be managed.

In plain words
What is it for?
Use them to start memory sessions, log messages, save important project details, and recall relevant past context.
Why use it?
They help preserve relevant decisions, preferences, and project context across conversations and reduce repeated explanations.

Cursor rule for Cursor

Install

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.

agentmods
npx agentmods add rules/firebolt-db/firebolt-memory-layer/cursorrules
Clone the repo
git clone --depth 1 https://github.com/firebolt-db/firebolt-memory-layer

Made for: Cursor.

Per session 333 This file is loaded in full into every session.
When invoked 333 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00333 $0.00333
Opus 5 $0.00167 $0.00167
Sonnet 5 $0.00067 $0.00067
Haiku 4.5 $0.00033 $0.00033

Measured yesterday against content hash bc62376c2d43, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cursorrules 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.

fml/.cursorrules · 34 lines

What it actually says

FML Memory System Rules

You have access to the FML (Firebolt Memory Layer) MCP server for persistent memory management.

Session Management

  • At the start of each conversation, call init_session with user_id="jon" to initialize or resume a memory session
  • Use the returned session_id for all subsequent memory operations

Automatic Conversation Logging

After EVERY user message, call add_to_working_memory with:

  • content_type="message"
  • A summary of the user's query and your response (keep it concise but capture key details)

When to Store to Long-Term Memory

Call store_memory when the user:

  • Makes decisions about architecture, tools, or approaches
  • Shares preferences about coding style, frameworks, or workflows
  • Provides important context about their project or environment
  • Asks you to remember something
  • Completes a significant task or milestone

When to Recall Memories

Call recall_memories or get_relevant_context when:

  • Starting work on a task (to get relevant background)
  • The user asks "what did we decide about X?"
  • You need context about the project, user preferences, or past decisions
  • The user references something from a previous conversation

Best Practices

  • Use get_relevant_context for complex queries - it automatically combines working + long-term memory
  • Pin important working memory items that shouldn't be evicted
  • Use entity tags (e.g., "table:users", "file:api.py") for better recall
  • Call checkpoint_working_memory at natural conversation boundaries or when the session gets long
Changes

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.

  1. yesterday First seen · 34 lines · 333 tokens per session scan A bc62376c2d43

Subscribe to this mod's changes

cursorrules is a cursor rule published in the GitHub repository firebolt-db/firebolt-memory-layer (2 stars, last pushed 6mo ago), licensed Apache-2.0. It adds 333 tokens to every session, about $0.0017 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.