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/sivang/sivan-claude-plugins/vertex-setupgit clone --depth 1 https://github.com/sivang/sivan-claude-pluginsWrote 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/commands/sivang/sivan-claude-plugins/vertex-setup)<a href="https://agentmods.dev/commands/sivang/sivan-claude-plugins/vertex-setup"><img src="https://agentmods.dev/badge/commands/sivang/sivan-claude-plugins/vertex-setup.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 | $0.00005 | $0.00448 |
| Opus 5 | $0.00003 | $0.00224 |
| Sonnet 5 | $0.00001 | $0.00090 |
| Haiku 4.5 | $0.00001 | $0.00045 |
Grade D, and why
vertex-setup scanned grade D with 2 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.
Reaches for credential fileshighPrivilege escalation
SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.
- ADC status: !`test -f ~/.config/gcloud/application_default_credentials.json && echo "ADC configured" || echo "ADC not configured"` Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- Existing config: !`cat ~/.claude/vertex-provider.json 2>/dev/null || echo "No config found"` What it actually says
Context
- Existing config: !
cat ~/.claude/vertex-provider.json 2>/dev/null || echo "No config found" - GCP project: !
gcloud config get-value project 2>/dev/null || echo "Not set" - ADC status: !
test -f ~/.config/gcloud/application_default_credentials.json && echo "ADC configured" || echo "ADC not configured"
Your Task
Guide the user through Vertex AI setup. Ask these questions ONE AT A TIME:
1. GCP Project ID
Ask for their GCP project ID. If gcloud shows a project, offer it as default.
2. Region
Ask which Vertex AI region:
- us-east5 (Columbus, Ohio) - Recommended
- europe-west1 (Belgium)
- asia-southeast1 (Singapore)
3. Authentication Method
Ask how to authenticate:
- ADC (Application Default Credentials) - Recommended for local dev
- Service Account - For CI/CD or servers
If ADC and not configured, guide them to run:
gcloud auth application-default login
If Service Account, ask for the path to the JSON key file.
4. Default Model
Ask which Claude model:
- claude-opus-4-5 (Most capable, $15/$75 per 1M tokens)
- claude-sonnet-4 (Balanced, $3/$15 per 1M tokens)
- claude-haiku-3-5 (Fast & cheap, $0.80/$4 per 1M tokens)
5. Save Configuration
After collecting all info, save with:
node "${CLAUDE_PLUGIN_ROOT}/lib/cli/setup.js" --project "PROJECT_ID" --region "REGION" --auth "METHOD" --model "MODEL"
6. Test Connection
Offer to test the connection:
node "${CLAUDE_PLUGIN_ROOT}/lib/cli/test-connection.js"
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 · 55 lines · 5 tokens per session scan D 4b416bc27e39
vertex-setup is a command published in the GitHub repository sivang/sivan-claude-plugins (1 stars, last pushed 7mo ago), licensed MIT. It adds 5 tokens to every session and 448 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it D with 2 findings (reaches for credential files, reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
ship
Branch, commit, open PR, gather Claude + every enabled AI reviewer (Copilot, CodeRabbit, etc.), fix/justify/resolve every finding, loop until clean, then merge. Run only when implementation is finished AND the owner has said to ship (e.g. "ship it") — never self-invoke just because the work looks done. To design and…
validate
Mechanically validate the sonu plugin repo before a PR — manifest sync, YAML frontmatter, shell-fence syntax, named-source and AI-attribution scans, cross-reference integrity, skill reachability. Only meaningful inside the claude-plugins repo; in any other repo, say so and stop.
build
Sequence decide → build → hand back for any feature or fix. Triage the change, run design-tree in plan mode (with you as the approval gate — skipped when /sonu:factory routes a ticket whose spec you already approved), build test-first under code-standards, then surface the riskiest parts before handing back to…
release
Prepare a release of the sonu plugin — decide the semver bump, sync the version and plugin description across all five homes, run /validate, then hand to /sonu:ship. Only meaningful inside the claude-plugins repo; in any other repo, say so and stop.
list
Show the project backlog (open items by default, sorted by priority then age). Ask for --all, --sort date or --priority high and Claude re-runs the engine with them.
prune
Delete old closed entries (status done/cancelled) left over from earlier plugin versions. Shows the list first; the deletion itself asks for your approval.