engage — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited engage (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Manage replies on your X posts to activate the 75–150× author reply conversation multiplier. This skill surfaces comments worth replying to, drafts high-quality replies, and posts them with your approval.
Why this matters: Author replies to comments create conversation threads — the single highest-weight signal in the Phoenix algorithm at 75–150× a like. A post with 5 author reply threads massively outperforms one with none.
The user provides one of:
mcp__claude-in-chrome__tabs_context_mcp(createIfEmpty: true)For a specific post:
mcp__claude-in-chrome__navigate(url: "<post_url>", tabId: <tab>)
mcp__claude-in-chrome__computer(action: "wait", duration: 3, tabId: <tab>)For notifications:
mcp__claude-in-chrome__navigate(url: "https://x.com/notifications", tabId: <tab>)
mcp__claude-in-chrome__computer(action: "wait", duration: 3, tabId: <tab>)Take a screenshot to verify state:
mcp__claude-in-chrome__computer(action: "screenshot", tabId: <tab>)If not logged in, stop and tell the user to log in manually.
On a specific post:
mcp__claude-in-chrome__get_page_text(tabId: <tab>)mcp__claude-in-chrome__javascript_tool(action: "javascript_exec", text: "
const articles = document.querySelectorAll('article');
const replies = Array.from(articles).slice(1).map((article, i) => {
const text = article.innerText;
const ariaLabels = Array.from(article.querySelectorAll('[aria-label]'))
.map(el => el.getAttribute('aria-label'));
return { index: i, text: text.substring(0, 300), meta: ariaLabels.slice(0, 5) };
});
JSON.stringify(replies.slice(0, 15), null, 2);
", tabId: <tab>)mcp__claude-in-chrome__computer(action: "scroll", coordinate: [640, 400], scroll_direction: "down", scroll_amount: 5, tabId: <tab>)
mcp__claude-in-chrome__computer(action: "wait", duration: 2, tabId: <tab>)On notifications page:
mcp__claude-in-chrome__get_page_text(tabId: <tab>)Categorize each comment by reply value:
High Priority (reply immediately):
Medium Priority (reply when time allows):
Low Priority (optional):
Present the prioritized list to the user:
Found [N] replies. Here are the ones worth responding to, ranked by conversation velocity impact:
>
1. [Username]: "[Reply preview]" — [why this is high priority] 2. [Username]: "[Reply preview]" — [why] ...
For each high-priority comment, draft a reply that:
Reply templates by comment type:
| Comment Type | Reply Strategy |
|---|---|
| Question | Direct answer + "curious, have you tried [related thing]?" |
| Disagreement | Acknowledge the point + share your reasoning + "what's your experience with [specific]?" |
| Personal experience | Validate + relate to your own experience + ask for detail |
| Feature request | Thanks + honest status/timeline + "what's the use case?" |
| Added context | Amplify their point + tag relevant people if appropriate |
Present drafted replies using AskUserQuestion for each high-priority comment:
AskUserQuestion:
"Replying to @username:
They said: \"[their comment]\"
My draft: \"[your drafted reply]\"
✅ 'post' — send this reply
✏️ Reply with changes — I'll revise (e.g., 'make it shorter', 'add a question')
⏭️ 'skip' — skip this one
❌ 'stop' — done replying for now"Handle the response:
For each approved reply:
mcp__claude-in-chrome__find(query: "reply button for this comment", tabId: <tab>)
mcp__claude-in-chrome__computer(action: "left_click", ref: "<ref_id>", tabId: <tab>)mcp__claude-in-chrome__computer(action: "wait", duration: 1, tabId: <tab>)mcp__claude-in-chrome__computer(action: "type", text: "<reply_content>", tabId: <tab>)mcp__claude-in-chrome__computer(action: "screenshot", tabId: <tab>)mcp__claude-in-chrome__find(query: "reply post button or submit reply button", tabId: <tab>)
mcp__claude-in-chrome__computer(action: "left_click", ref: "<ref_id>", tabId: <tab>)mcp__claude-in-chrome__computer(action: "wait", duration: 2, tabId: <tab>)If the user wants to build visibility before posting their own content:
mcp__claude-in-chrome__navigate(url: "https://x.com/search?q=[niche_topic]&f=live", tabId: <tab>)After the engagement session, summarize:
If the user says "dry run" or "just draft":
/engage again without dry-run to post the approved draftsfind for "reply text input"x.com/notifications/mentions~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.