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/container-registry/harbor-next/claude-mdgit clone --depth 1 https://github.com/container-registry/harbor-nextWrote 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/instructions/container-registry/harbor-next/claude-md)<a href="https://agentmods.dev/instructions/container-registry/harbor-next/claude-md"><img src="https://agentmods.dev/badge/instructions/container-registry/harbor-next/claude-md.svg" alt="Measured on agentmods" 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 | $0.00531 | $0.00531 |
| Opus 5 | $0.00266 | $0.00266 |
| Sonnet 5 | $0.00106 | $0.00106 |
| Haiku 4.5 | $0.00053 | $0.00053 |
Grade A, and why
harbor-next CLAUDE.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 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 — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Harbor Next
Enhanced fork of goharbor/harbor. Go backend in src/, Angular frontend in src/portal/, build automation via Taskfile.
Commands
task build # Build Go binaries
task test:quick # API lint + unit tests
task test:ci # Full CI pipeline
task images # Build/push Docker images
task dev:up # Local dev with hot reload
task apply-patches # Apply branches listed in taskfile/commercial-patches with jj
task release-ready # Verify those patches against a clean clone
Go test/build need generated API first: task build:gen-apis.
For an unsigned local release build and push, authenticate the container
engine first, then run task release-images-local RELEASE_VERSION=X.Y.Z IMAGE_TAG=<tag>. It defaults to 8gears.container-registry.com/8gcr-pr and
builds both linux/amd64 and linux/arm64.
PRs
- Branch off
main, never push direct. - Conventional Commits, capitalized subject:
feat: Add Foo,fix: Resolve Bar,upstream: Cherry-Pick Harbor Fix. - DCO sign-off required:
git commit -s. - Squash and merge only — other merge types break release-please.
- No
Co-Authored-By/ AI attribution trailers. - New features (
feat:) must add a## Release Notessection to the PR description. Its prose is extracted and rendered under## Highlightson the GitHub Release. See CONTRIBUTING.md → "Adding Release Notes to Your PR".
Release-please
main uses always-bump-minor; VERSION on main tracks the next development release while .release-please-manifest.json tracks the published release. release-X.Y branches use patch-only versioning. ci:, build:, chore:, test: are hidden from release notes.
exclude-paths: changes touching only .github/, docs/, tests/, or taskfile/ don't bump version — use ci: for CI-only changes.
Registry
Default: 8gears.container-registry.com/8gcr/. Override with REGISTRY_ADDRESS / REGISTRY_PROJECT. Publishing needs REGISTRY_USERNAME / REGISTRY_PASSWORD secrets.
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 · 42 lines · 531 tokens per session scan A 94bbf02d9692
harbor-next CLAUDE.md is an instructions file published in the GitHub repository container-registry/harbor-next (30 stars, last pushed today), licensed Apache-2.0. It adds 531 tokens to every session, about $0.0027 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 2026-09-03.
Other instructions, from other repositories
zebbern-kali-mcp CLAUDE.md
Claude Code instructions for zebbern/zebbern-kali-mcp, covering claude.md, two release tracks, and they are not the same, the local backend image goes stale silently, targeting the host from inside the container and releasing.
rosetta republish-assets-not-npm.instructions.md
Instructions for tikoci/rosetta: The republishassets workflow input is for GitHub Release assets and OCI tags only.
appwrite AGENTS.md
AGENTS.md instructions for appwrite/appwrite, covering appwrite, commands, stack, layout and libraries.
planning-with-files AGENTS.md
AGENTS.md instructions for OthmanAdi/planning-with-files, covering agents.md — planning-with-files agent reference card, commit rules, release checklist (12 steps), version bump scope and changelog format.
openrouter-mcp-multimodal AGENTS.md
AGENTS.md instructions for stabgan/openrouter-mcp-multimodal, covering agent instructions, before you ship, releasing (read this before publishing), short version and version files (must all match package.json).
OpenMicro AGENTS.md
Instructions for stephenleo/OpenMicro, covering openmicro — project conventions and releases.