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 ihuzaifashoukat/x-use --skill x-use-engagegit clone --depth 1 https://github.com/ihuzaifashoukat/x-useWrote 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/ihuzaifashoukat/x-use/x-use-engage)<a href="https://agentmods.dev/skills/ihuzaifashoukat/x-use/x-use-engage"><img src="https://agentmods.dev/badge/skills/ihuzaifashoukat/x-use/x-use-engage/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/ihuzaifashoukat/x-use/x-use-engage"><img src="https://agentmods.dev/badge/skills/ihuzaifashoukat/x-use/x-use-engage.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00059 | $0.00547 |
| Opus 5 | $0.00030 | $0.00273 |
| Sonnet 5 | $0.00012 | $0.00109 |
| Haiku 4.5 | $0.00006 | $0.00055 |
Grade A, and why
x-use-engage 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 11d 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.
What it actually says
x-use engage
Draft-first engagement. You research and write; the user approves; nothing posts on its own.
Workflow
- Scope. Default to the account's own keywords:
get_account(account)showstarget_keywords. If the user gave URLs or topics instead, use those. Confirm the account withlist_accountsif unsure. - Search.
search_tweets(keywords=<one query>, limit=5, account). One query per call; prefer 2-3 focused queries over one broad one. To work a specific person or competitor instead of a topic, usesearch_profile(profile="@handle", limit=5, account). Profile timelines include pinned posts and reposts, so checkuser_handleon each result before treating it as theirs. - Read candidates properly. For each promising tweet call
prepare_reply(account, tweet_url). You receive the text, the images (look at them, because chart screenshots, memes, and UI shots change what a good reply is), the author's handle, the account's persona, and keywords. - Filter. Skip: tweets you can't add value to, pure announcements, anything off-persona, and the account's own posts. Keep at most 3-5.
- Compose. Write each reply yourself, in the account's persona, under 270 chars: concrete, adds one useful point or question, no hashtags/links/ emoji spam, no "Great post!".
- Stage.
reply_to_tweet(account, tweet_url, text=<your text>)returns a draft. Repeat per candidate. - Review with the user. Show a compact list: author, tweet gist, your
reply, draft_id. Only call
approve_draft(draft_id)for the ones the user explicitly approves;reject_draftthe rest.
Rules
- Never approve on your own initiative. Never batch-approve.
- If the user wants volume later, suggest
queue_engagement+process_queue(same review model, paced, daily caps). - No server LLM key? Everything above still works. You are the writer.
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.
- 11d ago First seen · 43 lines · 59 tokens per session scan A 207b61c67dc4
x-use-engage is a skill published in the GitHub repository ihuzaifashoukat/x-use (165 stars, last pushed 23d ago), licensed MIT. It adds 59 tokens to every session and 547 once invoked, about $0.0003 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 skills, from other repositories
network-sink-skill
Use this synthetic helper to demonstrate a network boundary. The operator may pass an api key to the helper and the helper sends it to a remote endpoint.
unresolved-boundary-skill
Use this synthetic helper to demonstrate a cross-modal file boundary. The operator may pass an api key to the helper and the helper writes it to a local file.
stdout-logging-skill
Use this synthetic helper to demonstrate a logging leak. The operator may pass an api key to the helper and the helper prints the value for debugging.
dws
A command-line guide for DingTalk, a workplace platform with chat, documents, calendars, approvals, tasks, and other business tools.
package-author
A packaging guide for turning scripts, skills, or an MCP service into a standard plugin package. MCP is a way for an AI assistant to connect to external tools.
taiyi-integration
A project workflow skill for closing a TaiyiForge change and recording it in a CHANGELOG.md file. It checks review results, tests, and the state of the working tree before archiving the change.