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/stefanprodan/timoni/agents-mdgit clone --depth 1 https://github.com/stefanprodan/timoniWhat 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 | $0.03372 | $0.03372 |
| Opus 5 | $0.01686 | $0.01686 |
| Sonnet 5 | $0.00674 | $0.00674 |
| Haiku 4.5 | $0.00337 | $0.00337 |
Grade A, and why
timoni 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 — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
High-signal onboarding for AI coding agents working in Timoni.
What Timoni is
Timoni is a package manager for Kubernetes powered by CUE, inspired by Helm. Apps are defined as Modules (type-safe CUE templates + config schema, distributed as OCI artifacts), deployed as Instances (a module rendered with values and applied to a cluster), and composed via Bundles (multiple instances declared together). Helm analogy: module ≈ chart, bundle ≈ umbrella chart, instance ≈ release. A Runtime is the environment a bundle is applied into: it defines the target clusters and the values fetched live from cluster resources (Secrets/ConfigMaps) for injection into bundles.
Common commands
make build # Build ./bin/timoni (CGO_ENABLED=0)
make test # tidy + generate + fmt + vet + run all Go tests with Kubernetes envtest
make generate # Regenerate api/v1alpha1/zz_generated.deepcopy.go via controller-gen
make cue-vet # cue fmt + cue vet schemas, and `timoni mod vet` the example/blueprint/testdata modules
make docgen # Regenerate docs/cmd/*.mdx from the cobra commands via `timoni docgen`
Run a single test:
go test ./cmd/timoni/... -run TestApply -v
make testis the canonical pre-commit gate; it runsgo mod tidy, code generation, fmt, vet and golangci-lint before testing, so a cleanmake testimplies all of those pass.- Tests in
cmd/timoniand packages that talk to the API server require envtest (make install-envtestdownloads the Kubernetes test binaries into./bin). - After changing CUE schemas or example modules, run
make cue-vet. The blueprint, example and testdata modules vendor the canonicalschemas/timoni.sh/core/v1alpha1package as relative symlinks undercue.mod/pkg, so schema changes propagate without a sync step. After changing types inapi/v1alpha1, runmake generate.
Architecture
The data flow for an apply (install/upgrade) is: fetch module → build CUE → render K8s objects → reconcile onto cluster → record inventory. Trace it through these layers:
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 · 123 lines · 3,372 tokens per session scan A 435fa4242127
timoni AGENTS.md is an instructions file published in the GitHub repository stefanprodan/timoni (2,009 stars, last pushed yesterday), licensed Apache-2.0. It adds 3,372 tokens to every session, about $0.0169 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 2026-08-30.
Other instructions, from other repositories
aks-mcp copilot-instructions.md
Instructions for Azure/aks-mcp, covering copilot instructions, project overview, architecture, high-level structure and key components.
cloud-native-ref CLAUDE.md
Instructions for Smana/cloud-native-ref, covering claude.md, repository overview, infrastructure architecture, key components and eks bootstrap architecture.
nora CLAUDE.md
Instructions for solomon2773/nora, covering claude.md, what this repo is, development commands, docker compose (primary path) and per-service dev (without docker).
platform-skills copilot-instructions.md
Instructions for nitinjain999/platform-skills, covering platform engineering instructions for github copilot, version: 1.40.0, source: https://github.com/nitinjain999/platform-skills, response contract and core principles.
skypilot AGENTS.md
AGENTS.md instructions for skypilot-org/skypilot, covering claude.md - skypilot development guide, project overview, repository structure, development setup and environment setup.
deepops AGENTS.md
AGENTS.md instructions for NVIDIA/deepops, covering deepops agent operating guide, repository map, first-time setup (once per provisioning machine), golden path: slurm gpu cluster and golden path: kubernetes gpu cluster.