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 skills/huntsdesk/ve-kit/bootstrapnpx skills add HuntsDesk/ve-kit --skill bootstrapgit clone --depth 1 https://github.com/HuntsDesk/ve-kitWhat 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 | $0.00105 | $0.01471 |
| Opus 5 | $0.00053 | $0.00736 |
| Sonnet 5 | $0.00021 | $0.00294 |
| Haiku 4.5 | $0.00011 | $0.00147 |
Grade E, and why
bootstrap scanned grade E with 3 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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
If none exist, WebFetch `https://raw.githubusercontent.com/HuntsDesk/ve-kit/main/01-BOOTSTRAP.md` and read from there. If WebFetch isn't available, tell the user to run `curl -fsSL https://raw.githubusercontent.com/Hunts Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
> "Your setup is complete. You can remove the staging directory now: `rm -rf .ve-kit/`" Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
If none exist, WebFetch `https://raw.githubusercontent.com/HuntsDesk/ve-kit/main/01-BOOTSTRAP.md` and read from there. If WebFetch isn't available, tell the user to run `curl -fsSL https://raw.githubusercontent.com/Hunts How it starts
The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/bootstrap — run the ve-kit BOOTSTRAP protocol
This skill runs the ve-kit project-setup protocol interactively. The protocol file (01-BOOTSTRAP.md) is written as executable steps, not just docs — you READ it and then DO the phases, asking the user questions as you go.
When to invoke this skill
- User types
/bootstrap - User says "set up my project with ve-kit" / "run the bootstrap" / "follow @01-BOOTSTRAP.md"
- User says "upgrade my ve-kit" / "update my .claude/ config to the latest ve-kit"
- User says "add the Vibe Board to this project" (run Phase 4 only)
- User says "add the worker layer" (point them at
03-VE-WORKER.md— the worker is a separate setup)
Step 1 — Locate 01-BOOTSTRAP.md
Check these paths in order, use the first one that exists:
.ve-kit/01-BOOTSTRAP.md— staged viainit.sh(most common for first-time users)docs/ve-kit/01-BOOTSTRAP.md— already inside a project that has ve-kit checked in~/github/ve-kit/01-BOOTSTRAP.md— user has a separate ve-kit clone
If none exist, WebFetch https://raw.githubusercontent.com/HuntsDesk/ve-kit/main/01-BOOTSTRAP.md and read from there. If WebFetch isn't available, tell the user to run curl -fsSL https://raw.githubusercontent.com/HuntsDesk/ve-kit/main/init.sh | bash first.
Step 2 — Detect fresh setup vs upgrade
- Fresh setup: no
CLAUDE.md, no.claude/directory. Run the normal Phase 0 → Phase 11 flow. - Upgrade mode:
CLAUDE.mdexists OR.claude/exists. Run the upgrade path documented at the END of01-BOOTSTRAP.md(the "When a user says 'upgrade @bootstrap.md'" section) — diff their existing config against the current ve-kit template, ask what to update, don't overwrite anything without confirmation.
Step 3 — Run the protocol
Follow 01-BOOTSTRAP.md literally. The phases are:
| Phase | What |
|---|---|
| 0 | Prerequisite check (gcloud, firebase, node, Docker optional, Claude Code CLI) |
| 1 | Project questions (name, description, language, GCP project ID, default branch, deny-list additions) |
| 2 | Create .claude/ + docs/ directory structure (the MCP server is cloned separately in Phase 4 from HuntsDesk/ve-vibe-board) |
| 3 | Write rule files: riper-cat.md, code-quality.md, documentation.md, git-workflow.md, agent-board.md |
| 4 | Set up Vibe Board (Firestore project, service account, MCP server, .mcp.json) |
| 5 | Install hooks: block-todowrite.sh, session-handoff.sh, post-compact-recovery.sh, review-gate.sh, stop-compliance-check.sh |
| 6 | Permissions + deny list (settings.json merged with hooks config) |
| 7 | Write CLAUDE.md using the user's Phase 1 answers |
| 8 | Write docs/README.md |
| 9 | Install code intelligence plugins (language-specific LSPs) |
| 10 | Install starter skills (plan, review, go, /review-* family, this /bootstrap skill) |
| 11 | Self-verify (check files exist, hooks executable, board connectivity, skills loaded) |
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 · 93 lines · 105 tokens per session scan E ed2034ab0c9b
bootstrap is a skill published in the GitHub repository HuntsDesk/ve-kit (2 stars, last pushed 26d ago), licensed MIT. It adds 105 tokens to every session and 1,471 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it E with 3 findings (downloads and executes remote code, recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
tmux
Remote-control tmux sessions for interactive CLIs by sending keystrokes and scraping pane output.
summarize
Summarize or extract text/transcripts from URLs, podcasts, and local files (great fallback for “transcribe this YouTube/video”).
foundry-hosted-agent-validation
Step-by-step process for validating a Python Foundry hosted agent sample (under python/samples/04-hosting/foundry-hosted-agents/) end to end — running it locally (native runtime and azd ai agent run) and after deploying it to an Azure AI Foundry project with azd. Use this when asked to validate a hosted agent sample.
python-feature-lifecycle
Guidance for package and feature lifecycle in the Agent Framework Python codebase, including stage meanings, feature-stage decorators, feature enums, and how to move APIs from one stage to the next.
python-development
Coding standards, conventions, and patterns for developing Python code in the Agent Framework repository. Use this when writing or modifying Python source files in the python/ directory.
foundry-config-setup
Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.