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.
git clone --depth 1 https://github.com/deepfusionlabs/deep-initWrote 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/rules/deepfusionlabs/deep-init/deepinit-lean)<a href="https://agentmods.dev/rules/deepfusionlabs/deep-init/deepinit-lean"><img src="https://agentmods.dev/badge/rules/deepfusionlabs/deep-init/deepinit-lean/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/deepfusionlabs/deep-init/deepinit-lean"><img src="https://agentmods.dev/badge/rules/deepfusionlabs/deep-init/deepinit-lean.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.01693 | $0.01693 |
| Opus 5.5 | $0.00677 | $0.00677 |
| Sonnet 5.5 | $0.00339 | $0.00339 |
| Haiku 4.5 | $0.00169 | $0.00169 |
Grade A, and why
deepinit-lean 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 22d 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.
This is a copy
95% identical to cursorrules — 4 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
DeepInit lean rules (Windsurf, always-on)
Canonical context lives in AGENTS.md + .ai/docs/ — this file is a generated projection of the lean tier; do not edit inside the DeepInit markers (regenerated each run).
kemal — Agent Context
Kemal is a lean, Sinatra-style HTTP web framework for the Crystal language: a global-scope DSL (get "/" {... }, ws, sse, before_all, use, mount) layered over a fixed HTTP::Handler middleware chain that a single process-wide Config singleton assembles and owns.
Architecture
A composable HTTP::Handler middleware pipeline driven by a Sinatra-style global DSL over a process-global Config singleton. The whole src/kemal/ tree is loaded by three glob requires in src/kemal.cr; files reference siblings by fully-qualified constants, not per-file imports. Components map to .ai/docs/components/.
Components (7)
- core — entry point + global run/stop lifecycle + the
Configsingleton that assembles the ordered handler chain + the top-level DSL + CLI/SSL. →.ai/docs/components/core.md - handlers — the
HTTP::Handlermiddleware links (init, exception, head, override-method, path, static-file, filter) + the base handler with only/exclude matching. →.ai/docs/components/handlers.md - router — route storage/matching/dispatch: the terminal
RouteHandler+ LRU route cache, the modularRouterDSL, and param parsing with body-size guards. →.ai/docs/components/router.md - websocket — real-time transports:
WebSocketHandler(Origin allow-list), the WS route wrapper, and Server-Sent Events. →.ai/docs/components/websocket.md - helpers — global view/response helpers, ECR templating macros, the exception hierarchy, and path/zip utilities. →
.ai/docs/components/helpers.md - logging — pluggable request-logging abstraction (deprecated in favor of stdlib
Log); the defaultRequestLogHandler+ legacy/no-op loggers. →.ai/docs/components/logging.md - ext — stdlib monkey-patches (NOT under
Kemal::): reopensHTTP::Server::Context(params/redirect/route lookup + typed@store) andHTTP::Server::Response. →.ai/docs/components/ext.md
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.
- 22d ago First seen · 50 lines · 1,693 tokens per session scan A 4892135f82d4
deepinit-lean is a cursor rule published in the GitHub repository deepfusionlabs/deep-init (6 stars, last pushed 22d ago), licensed MIT. It adds 1,693 tokens to every session, about $0.0068 per session on Opus 5.5. A static security scan graded it A with 0 findings. It is 95% identical to cursorrules, differing in 4 lines, and is treated as a copy.
Other cursor rules, from other repositories
agent-context-doctor
Agent Context Doctor project rule adapter.
project-memory
This rule guides recording project events and decisions in projectmem.
cursor-rules
You have access to a persistent cross-session memory via the memory MCP server. Knowledge survives between sessions and is shared across agents working on the same project.
decision-logger
Two-layer memory architecture for board meeting decisions. Manages raw transcripts (Layer 1) and approved decisions (Layer 2). Use when logging decisions after a board meeting, reviewing past decisions with /cs:decisions, or checking overdue action items with /cs:review. Invoked automatically by the board-meeting…
archcore-context
Archcore knowledge base context — document types, MCP tools, and conventions for working with .archcore/ documents.
memento-conventions
Memento OS — reasoning artifact conventions for persistent AI memory.