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 instructions/gerardogrisolini/zencode/agents-mdgit clone --depth 1 https://github.com/gerardogrisolini/ZenCODEWhat 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 | $0.01084 | $0.01084 |
| Opus 5 | $0.00542 | $0.00542 |
| Sonnet 5 | $0.00217 | $0.00217 |
| Haiku 4.5 | $0.00108 | $0.00108 |
Grade A, and why
ZenCODE AGENTS.md 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 — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Project
- ZenCODE is a Swift 6.3 package whose main executable is
zen; it runs on macOS, Linux, and WSL with remote model providers. Package.swiftis authoritative for the root products, target dependencies, platforms, and build settings. The packages underSources/Featureshave their own manifests and are outside the root graph.Docs/architecture.mdis the compatibility and dependency-direction contract. Preserve public modules, executable names, wire formats, persisted formats, feature identities, and task-graph ownership unless an explicit migration changes them.- Update
CHANGELOG.mdwhenever a change is user-visible or otherwise warrants a release note. - Composition starts at
Sources/zen/CLI/ZenCODEMain.swift; shared session coordination is centered onAgentCoreSessionRunner, backend state is behindAgentCoreBackend, and direct-tool dispatch is behindDirectToolExecutor.
Build and Validation
- Every root Swift target enables
MemberImportVisibility; source files must import the modules that define the members they use. - Fast shared-runtime compile:
swift build --target ZenCODECore. - Use
swift test --filterwith the affected Swift Testing suite or test name. The full non-live suite isswift test; the CI and release gates useswift test --no-parallelfor process-global lifecycle coverage. Validate Linux behaviour in the container machine described under Linux Validation. - For memory-engine changes, run
swift test --filter MemoryEngineTests; memory-facade changes additionally exerciseswift test --filter MemoryServiceTestsandswift test --filter MemoryEnhancementTestsinZenCODECoreTests. - Main release product:
swift build -c release --product zen. - Root commands intentionally do not build optional feature executables. For an optional-feature change, run
swift test --filter BundledFeatureCatalogParityTests, validate the package from its own directory according to itsPackage.swift, and check its installed executable's--list-toolsoutput. - For sensitive-manifest changes, run
swift test --filter SensitiveManifestPermissionsTests; add the affected provider or setup suites when their persistence flows change. Package.resolvedis tracked and pins CI/release dependencies. Do not update it incidentally; for an intentional dependency change, runswift package resolve, review the exact pins, and include the lockfile.- Finish source changes with
git diff --check. Validate script edits withbash -n Scripts/*.sh; network, provider, and installer executions are dedicated checks rather than routine validation. - Release tags follow
Docs/release.md: strictvX.Y.Z, matchingZenPackageMetadata.version, with the same immutable tag or full commit SHA passed through--ref.
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 · 42 lines · 1,084 tokens per session scan A fd01e6e7cde7
ZenCODE AGENTS.md is an instructions file published in the GitHub repository gerardogrisolini/ZenCODE (5 stars, last pushed 2d ago), licensed MIT. It adds 1,084 tokens to every session, about $0.0054 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 instructions, from other repositories
FylloCode AGENTS.md
Instructions for Fioooooooo/FylloCode, covering agents.md, 项目概述, 技术栈, 目录结构 and 常用命令.
Mitsuro AGENTS.md
Instructions for honeycomb-Technologies/Mitsuro, covering agents guide: mitsuro, purpose, agents strategy, product language and compatibility and core architecture.
Mitsuro CLAUDE.md
Instructions for honeycomb-Technologies/Mitsuro: AGENTS.md is the canonical repository engineering guide for Mitsuro. Read and follow it in full before making changes. This compatibility file intentionally contains no separate rules so development policy cannot drift between agent harnesses.
FylloCode CLAUDE.md
Instructions for Fioooooooo/FylloCode, a project described as: FylloCode 是一个开源桌面应用,把 Claude Code、Codex 等本地 Coding Agent 接进项目工作流。派任务、审方案、追溯每一次改动,积累下来的规范和结论,换一个 Agent 也能继承。.
teaql-agent-kit AGENTS.md
AGENTS.md instructions for teaql/teaql-agent-kit, covering agents.md — teaql agent kit and hard requirements.
zot AGENTS.md
Instructions for patriceckhart/zot, covering working agreement for zot, product intent, starting a task, code ownership map and correctness contracts.