القائمة
Compose and send an email — using native GWS tools (never gog CLI for replies)
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | send-email |
| description | Compose and send an email — using native GWS tools (never gog CLI for replies) |
Compose and send an email from the configured account using native GWS tools.
Replying to an existing thread? → gws_gmail_reply(thread_id, body)
Sending a brand-new email? → gws_gmail_send(to, subject, body)
exec + gog gmail send? → NEVER for replies. BANNED.
thread_id from the task body, email search, or prior context.gws_gmail_reply(thread_id, body) — it handles reply-all recipient resolution, threading headers (In-Reply-To, References), and duplicate prevention automatically.cc for additional recipients beyond those already in the thread.gws_gmail_send(to, subject, body).content_type: "html" for HTML formatting (use <br> for line breaks, <p> for paragraphs).cc for CC recipients.gws_gmail_search(query="in:sent", max_results=1).gws_gmail_get(message_id=...) on the result.To header contains actual recipient addresses.To is empty — the send failed silently. Resend using gws_gmail_reply or gws_gmail_send with explicit recipients.log_interaction(contact_name="<NAME>", channel="email", direction="outgoing", content_summary="<SUMMARY>")
parent_task_id when spawning the email-responder so the child receives the task objective and constraints.gog gmail send --reply-all is BROKEN. The CLI's --reply-all flag silently resolves zero recipients and sends with an empty To field. This has caused real incidents. NEVER use exec + gog gmail send for replies. Use gws_gmail_reply instead.To field. If verification shows an empty To, the email went to nobody (except possibly CC). Resend immediately.gws_gmail_reply needs the thread_id, not the message_id. Get it from gws_gmail_search or gws_gmail_get(thread_id=...).Generate and apply structured code changes to a file
Fix agents that read stale production data instead of classifying direct benchmark prompts — add format detection at top of instructions
Diagnose and fix sub-agent configuration issues in YAML + instruction files that cause timeouts, tool failures, or degraded performance
Fallback persistence paths when primary tools (store_memory, create_note, write_file) are unavailable — memory blocks, task resolution fields, CRM notes, and reporting, in priority order.
Create, advance, and resolve CRM tasks correctly — respecting status transitions, handling retries, and avoiding common API pitfalls.
Audit and calibrate benchmark suite patterns against actual agent output to eliminate false-positives and false-negatives