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/zoverions/axiom-mesh/cursorrulesgit clone --depth 1 https://github.com/Zoverions/AXIOM-MESHWhat 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 | $0.00808 | $0.00808 |
| Opus 5 | $0.00404 | $0.00404 |
| Sonnet 5 | $0.00162 | $0.00162 |
| Haiku 4.5 | $0.00081 | $0.00081 |
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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AXIOM-MESH Agent Instructions
Capability is not authority. Discovery is not permission. Connection is not permission.
This file is the concise machine-facing operating guide for coding agents and automated contributors working with AXIOM-MESH. It does not grant repository, runtime, deployment, merge, or production authority.
Installation
Use the supported root setup command:
npm run setup
The supported Node.js line is pinned by .node-version and the package engine declarations. Do not substitute an unreviewed runtime or dependency tree merely because it works locally.
Configuration
Start from .env.example when local configuration is required. Treat credentials, tokens, provider configuration, production resolver mappings, and external destinations as authority-bearing material rather than ordinary convenience settings.
The supported privileged-effect path is:
Gateway -> Hypervisor -> Sandbox -> Grid
Do not create a parallel authority path around that sequence. Agent runtimes, adapters, discovery protocols, credentials, and reachable tools are clients or capabilities; they are not independent sources of permission.
Usage
Before changing supported behavior:
- Read
README.mdandAGENT-ENTRY.md. - Read
CONTRIBUTING.mdfor repository authority and contribution rules. - Inspect
mesh/config/capabilities.jsonbefore changing a capability claim. - Inspect
docs/security/CURRENT-BUILD-THREAT-MODEL.mdfor the current trust boundary. - Run narrow relevant checks while iterating.
- Run the full supported validation before presenting the change as ready:
npm run check npm run release:verify
For a read-only pre-action authority assessment, see agent-skills/axiom-authority-auditor/SKILL.md.
Authority rules for automated contributors
- A request to draft or propose a change is not permission to merge it.
- Repository write access is not production authority.
- A passing test is evidence about the tested claim; it is not production promotion.
- Do not activate providers, egress, remote execution, public endpoints, credentials, production resolver mappings, or deployment surfaces unless the governing change explicitly authorizes that activation.
- Do not weaken a failing safety or documentation boundary merely to make a proposed change pass.
- Missing, ambiguous, stale, or conflicting authority for a consequential external effect must not be converted into permission.
- Sensitive security findings belong in the process described by
SECURITY.md, not in a public proof-of-concept.
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.
- 2d ago First seen · 76 lines · 808 tokens per session scan A f1b308f0e817
cursorrules is a cursor rule published in the GitHub repository Zoverions/AXIOM-MESH (2 stars, last pushed 2d ago), licensed Apache-2.0. It adds 808 tokens to every session, about $0.0040 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
infra-deploy
ZizkaDB infrastructure, Docker Compose, deployment, CI/CD, and scripts.
testing
ZizkaDB testing conventions — per-layer test commands, integration test setup, and test file map.
backend-dashboard-contract
Backend contract the ZizkaDB dashboard depends on — invariants to preserve and KB to keep in sync.
coding-standards
ZizkaDB cross-cutting coding standards — always apply on every task.
enterprise-page-knowledge-base
Enterprise marketing page — copy guardrails, lead capture, QA, and cross-links.
sdk-integrations-mcp
ZizkaDB SDK (Python + TypeScript), integrations (LangChain/CrewAI), MCP server, and examples.