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/agentkitai/lore/cursorrulesgit clone --depth 1 https://github.com/agentkitai/loreWhat 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 | $0.00279 | $0.00279 |
| Opus 5 | $0.00139 | $0.00139 |
| Sonnet 5 | $0.00056 | $0.00056 |
| Haiku 4.5 | $0.00028 | $0.00028 |
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.
What it actually says
Lore — Persistent Memory (MCP)
You have access to Lore, a persistent memory system, via MCP tools.
Session Start
Call recent_activity(hours=24) at the beginning of each session to load recent context.
During Work
- Before debugging something that feels familiar:
recall("describe the problem") - After solving a non-obvious bug:
remember("problem + solution", type="lesson") - Key decisions or preferences:
remember("what was decided", type="note") - Check knowledge graph:
graph_query("topic")orentity_map("name")
Auto-Compaction Protection
Lore automatically snapshots your session when context grows large (30k chars).
You can also manually save state: save_snapshot(content="current state summary")
Memory Types
Use type parameter with remember: lesson, fact, preference, pattern, convention, note.
Key Tools
recall(query)— semantic search across all memoriesremember(content, type)— save something worth keepingrecent_activity(hours)— recent session contextsave_snapshot(content)— checkpoint current statetopics()— browse auto-generated topic hubsgraph_query(query)— explore knowledge graphon_this_day()— memories from this date in past years
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.
- yesterday First seen · 29 lines · 279 tokens per session scan A 4348ec9b2f4e
cursorrules is a cursor rule published in the GitHub repository agentkitai/lore (7 stars, last pushed 2d ago), licensed MIT. It adds 279 tokens to every session, about $0.0014 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 cursor rules, from other repositories
tutorial-system-guide
This guide documents the complete tutorial infrastructure in Windmill's frontend, enabling developers to create new interactive tutorials without re-exploring the codebase.
memory-informed-longitudinal-work
Long-running multi-session work (research, eval loops, iterative benchmarks) — resume prior lessons, capture per-run outcomes, build up stable truths over time.
session-memory
Use at conversation wrap-up or when the user explicitly indicates end-of-session — capture residual lessons not captured in-flight.
memory-informed-refactor
Use before substantive refactors — load relevant prior context, capture refactor insights as they land.
stock-data
Stock data domain (TW/US markets, FinMind, Massive, ticker insights, charts, translations). Auto-attached when editing stock-related code.
content-pipeline
Content/agent pipelines tier (podcast + news ingestion, transcription, summarization, ticker-sentiment, contentbuilder, wikibuilder). Auto-attached when editing pipelines/ code.