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/ibrahimhajjaj/wu-cli/groupsgit clone --depth 1 https://github.com/ibrahimhajjaj/wu-cliWhat 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 | $0.00007 | $0.00241 |
| Opus 5 | $0.00003 | $0.00120 |
| Sonnet 5 | $0.00001 | $0.00048 |
| Haiku 4.5 | $0.00001 | $0.00024 |
Grade A, and why
groups 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.
What it actually says
List WhatsApp groups using the wu MCP server.
By default wu_groups_list returns ALL known groups (the discovery surface) so the user can see JIDs to opt into via wu config allow <jid>. The constraint field tells you the current state: full (read+write), read (read-only), or none (not opted in).
Render the result so the user can see community structure:
- Group rows with
is_community: trueare community parents. - Rows with
linked_parent: <parent-jid>are subgroups; indent them under their parent. - Rows where
is_community_announce: trueare the community's announcement channel — call them out separately. - Plain groups (no
linked_parent, not a community) go after the community trees.
If the user asks for "groups I'm in" without further qualification, show everything. If they ask for "groups Claude can read or send to," pass allowed_only: true.
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 · 17 lines · 7 tokens per session scan A e7e31586af95
groups is a command published in the GitHub repository ibrahimhajjaj/wu-cli (14 stars, last pushed 19d ago), licensed MIT. It adds 7 tokens to every session and 241 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 2026-08-30.
Other commands, from other repositories
strict-development-rules
These rules apply strictly to all code modifications and additions within the Gemini CLI project.
serve
Start the main 1MCP runtime with 1mcp serve and use it for CLI mode, direct HTTP MCP clients, and template-aware runtime behavior.
proxy
Use 1mcp proxy as the maximum-compatibility bridge to a running 1MCP HTTP runtime.
admin
Manage CLI Admin sessions for Runtime Target Contexts.
cli-setup
Use the cli-setup command to install 1MCP startup docs and hooks for Codex or Claude in global or repo scope.
instructions
Use the instructions command to print the CLI playbook and current server inventory for AI agents using a running 1MCP serve instance.