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 skills add bengous/claude-code-plugins --skill startgit clone --depth 1 https://github.com/bengous/claude-code-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/skills/bengous/claude-code-plugins/start)<a href="https://agentmods.dev/skills/bengous/claude-code-plugins/start"><img src="https://agentmods.dev/badge/skills/bengous/claude-code-plugins/start/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/bengous/claude-code-plugins/start"><img src="https://agentmods.dev/badge/skills/bengous/claude-code-plugins/start.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00099 | $0.01724 |
| Opus 5.5 | $0.00040 | $0.00690 |
| Sonnet 5.5 | $0.00020 | $0.00345 |
| Haiku 4.5 | $0.00010 | $0.00172 |
Grade A, and why
start 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 17d 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 — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Plan at the frontiers
Planning writes the plan and its artifacts, nothing else. No code edit until the reviewer approves the plan. A throwaway prototype lives in the working directory and is not kept.
Explore, then let the reviewer pick each next step: settle the choices, draw a screen, try a throwaway, or plan. The plan is written to plan.md at the root of the working directory named by the Working directory: line at the end of this skill, and its artifacts live beside it; without that line, both go to plans/<date>/<slug>/ from the repository root unless the project names another place. Load one reference at a time, when its step starts.
1. Propose the next step
One sentence describes the diff: no plan. Say so, propose /vellum:stop, and implement once the reviewer runs it.
Otherwise explore, then propose the next step with mcp__vellum__propose, never by questions in the terminal: the reason in one sentence, the moves you offer, and the index of the one you recommend. The call waits while the reviewer picks in the review page, and returns their pick: "Accepted: ." for the one you recommended, "Chose: ." for another, one of theirs included, "Own: ." for their own words, "Declined." or "Declined: ." when they want none of the moves: propose others, in the light of the note. Take the step picked, then propose again once it is done. The moves:
grill: choices are open and change the architecture, an interface or the scope;choicesare their titles. Step 2.mockup: a screen words cannot settle;screennames it.references/visual.mdprototype: a question only a throwaway answers;questionnames it. It lives in the working directory; one that needs the codebase goes through/vellum:stop, and the plan comes in a fresh session from what it taught.plan: nothing left open. Step 3. Onceplan.mdexists the plan step is done: a proposal that offers it is refused, so revise the plan instead.
AskUserQuestion is refused while vellum is live. A reviewer may also pick a step on their own; it arrives as a prompt in the same words.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 17d ago Changed · +3 lines db74a6ac7791
- 21d ago Changed · +9 lines 75f854234095
- 23d ago Changed 3392aa618ce6
- 24d ago First seen · 54 lines · 99 tokens per session scan A 782cf0e88e3d
start is a skill published in the GitHub repository bengous/claude-code-plugins (5 stars, last pushed today), licensed MIT. It adds 99 tokens to every session and 1,724 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-09-23.
Other skills, from other repositories
html-ppt-hermes-cyber-terminal
OpenDesign + BYOK: choosing and wiring your own model, hands-on — cost, quality, and the routing decision. Built as a decision-grade AI literacy deck for engineers, IT, applied-AI teams.
development
An index of programming guidance for Python, Go, Rust, TypeScript, Java, C++, and shell scripting.
tencent-docs
A Tencent Docs assistant for creating, reading, editing, organizing, importing, and exporting online documents, spreadsheets, presentations, diagrams, and forms.
wechat-search
Search WeChat Official Account articles using OpenClaw's web search, Tavily API, and web fetch capabilities with compliance-focused design.
debugging-executions
Debug failed or wrong-output workflow executions using executions tools. Load when the user reports execution failures, unexpected node output, empty parameter values after a successful run, or a node showing a red or failed expression error. Also load when a live run you started for the user fails with an error from…
flow-next-tracker-sync
Project a flow-next spec to a tracker issue (Linear, GitHub, GitLab, Jira) and reconcile two-way. Use when asked to sync to a tracker. NOT plan-sync.