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/livetrack-x/spec-driven-ai-development/cursor-mini-sdadgit clone --depth 1 https://github.com/LiveTrack-X/spec-driven-ai-developmentWhat 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 | $0.01300 | $0.01300 |
| Opus 5 | $0.00650 | $0.00650 |
| Sonnet 5 | $0.00260 | $0.00260 |
| Haiku 4.5 | $0.00130 | $0.00130 |
Grade A, and why
cursor-mini-sdad 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.
How it starts
The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Mini SDAD Agent Rules
Use this file when the project is too small for the full SDAD Protocol but still needs basic protection against scope drift, vague completion, or session context loss.
Core 5
Current beats historical. Evidence beats confidence. Active beats interesting. Owner decision beats AI momentum. Repeated pain becomes a rule. Compression first. Gates stay real.
For repeated pain or one high-risk control failure, capture the root cause and smallest durable control plus regression evidence; later Keep/Refine/Merge/Retire it. Prefer refining an existing rule, flow, check, or template.
A current applicable owner instruction overrides older direction. Stop affected work and restate the unit before continuing; review/reference intent stays read-only.
Scale
This project uses Mini SDAD.
Do not create the full SDAD folder structure unless the owner asks or the work grows beyond this file.
Escalate to Standard or Full SDAD only when:
- three or more persistence/evidence signals now apply across future sessions, owner return, and evidence beyond AI confidence,
- durable TODOs or review findings no longer fit this one-file control surface,
- multiple AI tools or reviewers become involved,
- a release, production, migration, destructive action, real user data, auth, money, security, rollback, or another owner-controlled gate appears.
Sensitive Data Boundary
Context size and data authorization are separate checks. Use metadata-only
inspection by default for secret-bearing or private inputs. Do not read, copy,
transmit, summarize, or paste .env files, credentials, private keys, tokens,
cookies, raw customer records, or private corpora into an AI context unless the
task requires it and owner policy plus tool policy explicitly permit it.
Prefer redacted samples, schemas, counts, filenames, and hashes. Keep sensitive material out of prompts, logs, TODOs, findings, handoffs, and generated artifacts. If authorization is unclear, stop before reading the content and ask.
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 · 159 lines · 1,300 tokens per session scan A c85ce5c64fd5
cursor-mini-sdad is a cursor rule published in the GitHub repository LiveTrack-X/spec-driven-ai-development (4 stars, last pushed 1mo ago), licensed MIT. It adds 1,300 tokens to every session, about $0.0065 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
lean
The smallest working change. Efficiency is not carelessness. Active during delivery in the aiops bundle; off during grill/alignment.
project-organization
Rules for organizing Electron projects with clear separation between foundational/template code and application-specific code. Template code goes in src/, app-specific code goes in src/app/.
software-factory
Software factory operating rules.
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.