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 commands/sprklai/zenii/shipgit clone --depth 1 https://github.com/sprklai/zeniiWhat 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 | $0.00000 | $0.02121 |
| Opus 5 | $0.00000 | $0.01060 |
| Sonnet 5 | $0.00000 | $0.00424 |
| Haiku 4.5 | $0.00000 | $0.00212 |
Grade A, and why
ship 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 3d 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 — 160 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Format, Update Docs, Scan Secrets, Commit & Push to Main
This command formats the code, updates documentation to reflect changes, scans for leaked secrets, commits, and pushes to the remote main branch. It will REFUSE to proceed if any secrets or tokens are detected. It runs fully automatically with NO user prompts unless secrets are found.
Instructions
Follow these steps strictly and sequentially. Stop immediately if any step fails. Do NOT ask the user for confirmation between steps — proceed automatically unless a secret is detected.
Step 1: Verify branch
Run git branch --show-current to confirm we are on the main branch.
If NOT on main, STOP and tell the user: "You are on branch X, not main. Switch to main first or confirm you want to push from this branch."
Step 2: Format the code
Run the following formatters:
cargo fmt --all
If a web/ directory exists with a package.json, also run:
cd web && bun run format 2>/dev/null || npx prettier --write . 2>/dev/null; cd -
Step 3: Check i18n
If a web/ directory exists with project.inlang/settings.json, run:
cd web && bun run check
This runs svelte-kit sync && svelte-check which validates paraglide i18n message keys, types, and compiled output. If there are errors (missing message keys, type mismatches, stale paraglide output), fix them before proceeding. Do NOT ask the user — just fix and re-run.
Step 4: Run lints
Run cargo clippy --workspace to catch lint issues. If there are warnings or errors, fix them before proceeding. Do NOT ask the user — just fix and re-run.
Step 5: Update documentation
Review the staged and unstaged changes (use git diff and git diff --cached) to understand what has changed since the last commit. Then update the following documentation files only if the code changes warrant it — do not make gratuitous edits:
README.md— Update if there are new features, removed features, changed commands, new dependencies, or altered project structure.CHANGELOG.md— Add entries for any user-facing changes (features, fixes, breaking changes). Follow the existing format and group under the current unreleased version section. Create a new section if needed.website-data.json— Sync derived metrics with the current codebase state:version: read fromgrep -m1 '^version = "' Cargo.toml | sed 's/.*"\(.*\)".*/\1/'totalEndpoints: count viabash scripts/count-endpoints.sh(readsdocs/api-endpoints.md— update that file first if routes changed)agentTools: count viagrep -c 'registry\.register\b' crates/zenii-core/src/boot.rsbackendTests: count viacargo test --workspace 2>&1 | awk '/test result/{sum+=$4} END{print sum}'- Other fields (
binarySize,debRpmSize,startupSeconds,llmProviders,messagingChannels,crates,securityLayers,platformTargets,compileFlags) — update manually only if the code changes warrant it. - Read the current
website-data.json, apply only the fields that changed, write back. - After writing
website-data.json, readbackendTestsfrom it and update the test count badge inREADME.md: replacetests-<N>-bluewithtests-<backendTests>-blueandalt="<N> tests"withalt="<backendTests> tests". - If
totalEndpointschanged, update only the route count in the GitHub repo description (preserve everything else):ENDPOINTS=$(jq '.totalEndpoints' website-data.json) CURRENT_DESC=$(gh repo view sprklai/zenii --json description -q '.description') NEW_DESC=$(echo "$CURRENT_DESC" | sed "s/[0-9]\+ API \(routes\|endpoints\)/${ENDPOINTS} API endpoints/") gh repo edit sprklai/zenii --description "$NEW_DESC"
docs/directory — Update relevant docs if the changes affect:docs/architecture.md— new modules, traits, data flows, or structural changesdocs/phases.md— phase status changes or deliverable updatesdocs/processes.md— process flow changesdocs/cli-reference.md— new or changed CLI commands/flags- Any other doc that describes behavior that has changed
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.
- 3d ago First seen · 160 lines · 0 tokens per session scan A ddcc2297ff4f
ship is a command published in the GitHub repository sprklai/zenii (23 stars, last pushed 24d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,121 tokens. 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 commands, from other repositories
code-push-hermit
Commit-and-push are the textbook grunt-work target: short, mechanical, deterministic, and unrelated to reasoning. This skill keeps the deciding parts in Claude and delegates the git commands to Hermit.
speckit.git.commit
Auto-commit changes after a Spec Kit command completes.
commit-all
Group all changes semantically and commit each group separately.
commit
Safe git commit practices for MindRoom.
commit
智能生成 Git 提交信息并提交.
commit
Create a Conventional Commit with signoff.