How to Write a LinkedIn Reply From Profile and Thread Context
Use AGNURA Write Reply on LinkedIn messaging: load profile and thread (best-effort), steer with What I want to say, review, insert manually, and send yourself.
You will be able to draft a LinkedIn messaging reply from recipient profile and conversation thread context, review it, insert it into compose, and send it yourself.
AGNURA AI Assistant adds Write Reply on LinkedIn messaging compose. It can auto-load profile and thread text (best-effort), call managed AI with your LinkedIn reply prompt and optional knowledge base, then wait for Insert into message. It never sends the LinkedIn message for you.
Hub map: LinkedIn Writing Workflows: Messages, Invitations, Comments, and Posts. If you only need to polish text already in the box, use Rewrite Message instead (draft-only; no profile/thread/KB).
Prerequisites
- Desktop Chrome, signed into LinkedIn in the same profile as AGNURA
- AGNURA installed; Activate LinkedIn on (Settings → General; default on for new installs)
- Signed in to AGNURA with AI credits remaining
- Non-empty LinkedIn reply prompt (Settings → LinkedIn Prompts)
- Optional: Knowledge base in General settings
- A LinkedIn messaging compose box open (new or existing chat)
Refresh the LinkedIn tab after changing Activate LinkedIn if the buttons do not appear.
Direct answer
- Open the LinkedIn conversation and click into the message box.
- Click Write Reply in the compose footer.
- Confirm Recipient profile and/or Conversation thread loaded (Edit if needed).
- Optional: fill What I want to say.
- Click Generate, edit the result, then Insert into message.
- Review in LinkedIn and click Send yourself.
Exact steps
1. Open messaging compose
- Go to LinkedIn Messaging.
- Open the chat (or start a new message where compose is available).
- Confirm the footer shows Voice to Text (mic), Write Reply, and Rewrite Message on the left side after the emoji control.
If those controls are missing: confirm linkedin.com, Activate LinkedIn is on, and the tab was refreshed after install or toggle changes. See AGNURA Not Showing or Not Generating: Troubleshooting.
2. Open Write Reply
- Click Write Reply.
- Wait for the modal fields to populate.
Nothing is sent to managed AI until you click Generate inside the modal.
3. Confirm profile and thread context
The modal typically includes:
| Field | Role |
|---|---|
| Reply prompt | Standing instructions from LinkedIn Prompts. Expand or Edit for this run if needed. |
| Recipient profile | Best-effort summary: name, headline, company, location, about, skills, experience, birthday from Highlights when shown, and top Activity posts when visible on the profile. |
| Conversation thread | Date-ordered transcript from the message list when available (may be empty for new chats). |
| What I want to say | Optional one-run note (not saved to Settings). |
How profile can load (best-effort):
- From an open profile page and/or messaging header
- When messaging only exposes a profile URL, AGNURA may use LinkedIn session-backed fetch paths (including internal profile APIs, a hidden iframe, and if needed a brief background profile tab scrape)
| Outcome | What to do |
|---|---|
| Profile and/or thread look right | Continue |
| Fields empty or thin | Open the person’s profile in another tab, return, reopen Write Reply; or Edit and paste what you need |
| Private / missing sections | Do not invent them; write from what you can see |
Write Reply requires at least one of recipient profile or conversation thread. Profile-only works for new conversations.
4. Generate the draft
- Click Generate (above the result area).
- Wait for Generated message (hidden until success).
- Edit the draft in the modal before insert.
What managed AI receives (plain text order):
- Brief instructions for how to use each section
- Reply prompt
- What I want to say (or none)
- Knowledge base (or none)
- Recipient profile
- Conversation thread (or no prior messages)
The model is steered to follow the prompt first, prioritize What I want to say when provided, personalize from profile data, respond to the thread when messages exist, and use the knowledge base for factual context only.
If the profile shows a birthday today, openings may lead with a birthday wish. Recent/belated birthday signals can steer a belated wish first when present. Still verify before you send.
5. Insert, then send yourself
- Click Insert into message (or Copy).
- Read the full LinkedIn compose box.
- Cut invented compliments, wrong employers, or claims you cannot defend.
- Click Send in LinkedIn only when you approve.
Insert is manual. Generate does not auto-insert or auto-send.
What success looks like
| Check | Expected |
|---|---|
| Controls | Write Reply visible on messaging compose |
| Context | Profile and/or thread present in the modal |
| Draft | Appears only after a successful Generate |
| Insert | Text lands in compose when you choose Insert |
| Send | You click Send in LinkedIn |
Common failures and fixes
| Problem | Fix |
|---|---|
| No Write Reply button | Activate LinkedIn on; refresh LinkedIn; confirm messaging compose is open |
| Generate fails / no key | Sign in and check credits at app.agnura.com; check model access/billing |
| Empty LinkedIn reply prompt | Set and Save under LinkedIn Prompts |
| Empty profile and thread | Open profile page; retry; Edit and paste; need at least one context source |
| Thin profile (no birthday/skills/experience) | Expected when LinkedIn hides or lazy-loads sections; do not invent |
| Draft ignores your point | Fill What I want to say; regenerate |
| Only need wording polish | Use Rewrite Message (draft + rewrite prompt only) |
When to escalate to human judgment
- Legal, pricing, or hiring commitments appear in the draft
- The thread includes confidential data you should not send to managed AI
- Tone is high-stakes (complaint, executive, sensitive personal topic)
- Profile data looks scraped incorrectly
In those cases, write more yourself or shorten what you send.
How AGNURA helps this job
| Need | With AGNURA |
|---|---|
| Collect profile + thread near compose | Best-effort auto-load into Write Reply |
| Steer one message | What I want to say + editable reply prompt |
| Keep standing facts handy | Optional knowledge base |
| Stay in control | Manual Insert; you Send |
Compared with common alternatives
| Approach | Trade-off |
|---|---|
| Copy profile/thread into a chatbot tab | Works; more paste friction |
| Rewrite Message only | Faster polish; no profile/thread/KB |
| Auto-DM tools | Volume; not AGNURA’s job; higher platform risk |
| AGNURA Write Reply | Same-tab context + review/insert on LinkedIn web |
Privacy note: when you click Generate, prompts, knowledge base, profile, and thread text go to AGNURA managed AI (credits). See AGNURA Privacy in Plain Language. Follow LinkedIn’s terms.
Related guides
- LinkedIn Writing Workflows: Messages, Invitations, Comments, and Posts
- How to Generate a Gmail Reply With Thread Context
- Generate vs Rewrite in Gmail: Which Context Is Used?
- Sign In and Use AGNURA AI Credits
- AGNURA Not Showing or Not Generating: Troubleshooting
Soft next step
Open one real LinkedIn chat, run Write Reply once, and force a full human edit before Send. Install AGNURA if you need the messaging controls:
Try AGNURA in Chrome
Draft in Gmail, LinkedIn, WhatsApp, and Reddit. You review every message before send.
Add to Chrome