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/kyopark2014/agent-plugins/startgit clone --depth 1 https://github.com/kyopark2014/agent-pluginsWhat 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 | $0.00008 | $0.01147 |
| Opus 5 | $0.00004 | $0.00574 |
| Sonnet 5 | $0.00002 | $0.00229 |
| Haiku 4.5 | $0.00001 | $0.00115 |
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 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 — 158 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Start Command
If you see unfamiliar placeholders or need to check which tools are connected, see CONNECTORS.md.
Initialize the task and memory systems, then open the unified dashboard.
Instructions
1. Check What Exists
Check the working directory for:
TASKS.md— task listCLAUDE.md— working memorymemory/— deep memory directorydashboard.html— the visual UI
2. Create What's Missing
If TASKS.md doesn't exist: Create it with the standard template (see task-management skill). Place it in the current working directory.
If dashboard.html doesn't exist: Copy it from ${CLAUDE_PLUGIN_ROOT}/skills/dashboard.html to the current working directory.
If CLAUDE.md and memory/ don't exist: This is a fresh setup — after opening the dashboard, begin the memory bootstrap workflow (see below). Place these in the current working directory.
3. Open the Dashboard
Do NOT use open or xdg-open — in Cowork, the agent runs in a VM and shell open commands won't reach the user's browser. Instead, tell the user: "Dashboard is ready at dashboard.html. Open it from your file browser to get started."
4. Orient the User
If everything was already initialized:
Dashboard open. Your tasks and memory are both loaded.
- /productivity:update to sync tasks and check memory
- /productivity:update --comprehensive for a deep scan of all activity
If memory hasn't been bootstrapped yet, continue to step 5.
5. Bootstrap Memory (First Run Only)
Only do this if CLAUDE.md and memory/ don't exist yet.
The best source of workplace language is the user's actual task list. Real tasks = real shorthand.
Ask the user:
Where do you keep your todos or task list? This could be:
- A local file (e.g., TASKS.md, todo.txt)
- An app (e.g. Asana, Linear, Jira, Notion, Todoist)
- A notes file
I'll use your tasks to learn your workplace shorthand.
Once you have access to the task list:
For each task item, analyze it for potential shorthand:
- Names that might be nicknames
- Acronyms or abbreviations
- Project references or codenames
- Internal terms or jargon
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 · 158 lines · 8 tokens per session scan A 9a7ca7c70f4e
start is a command published in the GitHub repository kyopark2014/agent-plugins (4 stars, last pushed 24d ago), licensed MIT. It adds 8 tokens to every session and 1,147 once invoked, about $0.0000 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 2026-08-31.
Other commands, from other repositories
status
Show a comprehensive overview of the current Zilliz Cloud cluster — context, cluster details, databases, and collections with stats.
improve-prompt
Analyze and enhance vague or incomplete prompts before execution.
commit
Create well-structured git commits with meaningful messages.
explain-code
Break down complex code into clear, understandable explanations.
refactor
Analyze code and suggest meaningful refactoring improvements.
review-code
Perform a thorough code review on the specified files or recent changes.