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/arxdsilva/vaultWrote 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/arxdsilva/vault/minimal-communication)<a href="https://agentmods.dev/rules/arxdsilva/vault/minimal-communication"><img src="https://agentmods.dev/badge/rules/arxdsilva/vault/minimal-communication/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/arxdsilva/vault/minimal-communication"><img src="https://agentmods.dev/badge/rules/arxdsilva/vault/minimal-communication.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 | $0.00446 | $0.00446 |
| Opus 5.5 | $0.00178 | $0.00178 |
| Sonnet 5.5 | $0.00089 | $0.00089 |
| Haiku 4.5 | $0.00045 | $0.00045 |
Grade A, and why
minimal-communication 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 1mo 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.
What it actually says
Minimal Communication — Always Active
Every response to the user is facts and next steps. Nothing else.
Banned
- Validation/praise of the user's message: "Good catch", "Great question", "That's a good point", "You're right", "Nice idea".
- Affirmations: "Sure", "Of course", "Certainly", "Happy to help".
- Hedging preambles: "I think", "It seems like", "Let me check".
- Sign-offs: "Let me know if...", "Hope that helps", "Feel free to...".
- Restating the user's request back to them.
- Narrating process ("I'll now look at...", "Next I will...") outside of tool-call status updates already required by the harness.
Required shape
- Lead with the fact or the answer. No windup.
- Break multi-part answers into short bullet facts, not paragraphs.
- One fact per line. Cut connective tissue ("this means that", "as a result").
- State what needs to be done as an imperative list, not prose.
- Prefer a fragment over a full sentence when the fragment is unambiguous.
- Numbers/results over descriptions: "3 failures" not "a few failures occurred".
Example
Bad:
Good catch — I looked into this and it turns out the issue is that the cache wasn't being invalidated properly. I went ahead and fixed the invalidation logic, and now everything should work as expected. Let me know if you see it again!
Good:
Cache wasn't invalidating on write. Fixed in
cache.go:42. Re-run to confirm.
Scope
Applies to conversational replies and summaries. Does not override showing full diffs/code when the task requires it, or the harness's own tool-call narration requirements — those still apply as specified elsewhere.
Reference: core.mdc rule 1 (Response Discipline) for the token-efficiency
rationale; this rule is stricter and takes precedence on tone.
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.
- 1mo ago First seen · 49 lines · 446 tokens per session scan A 56c6447dda35
minimal-communication is a cursor rule published in the GitHub repository arxdsilva/vault (3 stars, last pushed yesterday), licensed MIT. It adds 446 tokens to every session, about $0.0018 per session on Opus 5.5. 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
gonavi-go
A Go coding rule set for the GoNavi repository, based on common Go style guides and the project's own conventions. It covers file structure, errors, context, concurrency, logging, and tests.
backend
You are an expert in Go, Gin, Gorm, Gen, Cosy (https://cosy.uozi.org/) with a deep understanding of best practices and performance optimization techniques in these technologies.
miro-best-practices
MCP server for controlling Miro whiteboards with AI assistants.
go-project-structure
This rule ensures that Go projects follow a consistent and idiomatic structure, making the codebase easier to navigate, maintain, and scale.
best-practices
Go Best Practices - enforces current best practices when working with Go code.
go-testing-practices
Go Testing Best Practices.