Extract the rules
Read docs/agent-policy.md and list the rules agents must follow, the one exception we allow and the cost the policy has for developers. Use only what the document states and mark anything you had to infer.
×Continue × LinkedInPlatform workflow
Continue / LinkedIn
Continue for LinkedIn: Governed Agent Workflows
Connect Continue to Tareno to turn an approved agent policy, its permitted tools and its review checkpoint into a LinkedIn post, approved by you in Tareno.
Short answer
Can Continue write a LinkedIn post about how our team governs its coding agents?
Yes, once Tareno is configured as an MCP server in Continue's agent mode. Continue can turn your approved policy, the tools agents may use and the review checkpoint into a LinkedIn draft and save it through Tareno. Tareno has not completed a client test for Continue, and Continue's MCP page does not mention OAuth. Scheduling needs your approval.
Last reviewed 2026-10-24 · Markdown version for AI agents
Continue for LinkedIn
Platform teams that let developers use coding agents usually write a policy: which tools an agent may call, which actions need a human and where the review sits. Those rules are exactly what other teams want to see, and they make a convincing LinkedIn post because they are specific. Continue is open source and configured in files you control, so the setup itself fits a team that cares about governance. This page follows a platform team publishing its agent policy, covering how Continue loads MCP servers, why it only works in agent mode and what to check when authentication is not OAuth.
Example: a platform team shares its agent policy
Imagine a platform team of six that wrote a one-page policy for coding agents: read-only by default, write access only through pull requests, and a human check before anything leaves the company. The lead asks Continue in agent mode to read the policy document and turn it into a 220-word LinkedIn post for engineering leaders. Continue lists the three rules, the one exception the team allowed and the cost of the policy, which was slower merges. It marks a sentence about "zero incidents" as unsupported by the document. The lead removes it, Continue confirms her profile with list_accounts and saves the draft through create_drafts, and she requests a Monday 08:30 slot that waits as a pending action.
Step by step
- 01
Create the MCP server file
Continue's documentation says to create a .continue/mcpServers folder at the top of your workspace and add a YAML file, or put an mcpServers block in the main config. Standalone block files need name, version and schema metadata.
- 02
Use a supported transport
The documentation lists stdio, sse and streamable-http as supported transports. Tareno's hosted endpoint is a remote server, so choose streamable-http and give it the endpoint from the Remote MCP setup page.
- 03
Check authentication
The page does not mention OAuth. If your Continue version cannot complete an OAuth sign-in for a remote server, Tareno also supports API keys for compatible clients, sent as a bearer header and kept in an environment variable, never in a file you commit.
- 04
Switch to agent mode
Continue's documentation says MCP can only be used in agent mode. Switch modes, then ask for
list_accountsas a read-only test; if your LinkedIn profile appears, the connection works. - 05
Give the policy and the cost
Share the policy document and ask Continue for the rules, the exception and the cost. A governance post that admits what the policy slowed down is more credible than one that lists only benefits.
- 06
Save, request and approve
Continue saves the post through
create_drafts. Give a date, a time and an IANA time zone, and approve the pending request in the Tareno dashboard.
Prompts to copy
Draft the post
Write a 220-word LinkedIn post for [engineering leaders] about our agent policy: read-only by default, writes only through pull requests, a human check before anything leaves the company. Include the cost, slower merges, and flag every claim that the policy document does not support.Save the draft
Use Tareno to confirm my LinkedIn profile [name] and save the approved text as a draft. Return the draft ID and the text, and do not request a publishing time until I give you the day and the time zone.Request the slot
Ask Tareno to schedule draft [draft ID] for [Monday 08:30] in the [Europe/Vienna] time zone. Return the pending action ID and remind me that I must approve it in the Tareno dashboard before it can publish.LinkedIn limits and tools in Tareno
- Text
- up to 3,000 characters; hashtags and mentions are supported; basic formatting is supported
- Images
- jpg, png, gif, up to 10 MB each, up to 20 per post, recommended ratios 1.91:1, 1:1
- Video
- mp4, mov, 3 to 10 min, up to 50 MB, one video per post, ratio 16:9, 1:1
- Links
- links can sit inside the text and get a preview card
- What Tareno can read
- posts published outside Tareno: no; comments: no; transcripts: no; video download: yes
As configured in Tareno on 2026-10-24. The live rules for your own account come from the get_platform_schema tool.
What LinkedIn itself documents
- Organic posts can contain text, images, videos, documents, articles, several images or polls. Organic carousels are not supported by the Posts API; they exist only as sponsored content.
- Posting for a company page requires one of the page roles administrator, direct sponsored content poster or content admin. Posting as a member needs the w_member_social permission.
- An article post needs its title, description and thumbnail set explicitly, because the API does not scrape the link for a preview.
- Reading a member's own posts needs r_member_social, a restricted permission that is available to approved users only.
Source: Microsoft Learn: LinkedIn Posts API, read on 2026-10-24. Platform rules change; check the page before you rely on a number.
LinkedIn in Tareno
- Tareno cannot read LinkedIn posts published outside Tareno or comments on them, so Continue cannot see how other teams described their agent policies.
- LinkedIn's documentation says posting as a member needs the w_member_social permission, granted when you connect your own profile in Tareno.
- Tareno's configuration allows up to 3,000 characters on LinkedIn, so a 220-word post can include a short list of rules and a link to the full policy.
- LinkedIn offers no transcripts through Tareno, so a recorded internal talk about the policy has to be summarized in writing first.
What does not work
- Continue cannot use the Tareno tools outside agent mode, so a plain chat will not show them however the server is configured.
- It cannot decide which parts of your policy are safe to publish, so a person has to read the draft before approval.
- It cannot publish by itself, since scheduling and publishing requests wait for your approval in the Tareno dashboard.
Mistakes to avoid
- Claiming that governance prevented incidents. Without a baseline, say what the rules require, not what they have avoided.
- Publishing the exact permission list of internal tools. Describe the principle, not the inventory, so the post does not map your systems.
- Forgetting agent mode. In plain chat the Tareno tools will not appear, which looks like a broken connection.
- Putting an API key into a committed YAML file. Use an environment variable, and keep the file free of credentials.
More Continue guides
LinkedIn with other agents
- OpenClaw: Can OpenClaw turn an article you message it into a LinkedIn post?
- Hermes Agent: Can Hermes Agent run a recurring LinkedIn plan for an expert?
- Grok Bot: Can Grok Bot draft a LinkedIn company update from a change you describe?
- Codex: Can Codex explain a technical change to LinkedIn readers?
- Claude Code: Can Claude Code write a technical case study for LinkedIn?
- Claude Cowork: Can Claude Cowork turn a white paper into LinkedIn thought leadership?
- Cursor: Can Cursor keep a LinkedIn product journal from your decisions?
- Google Antigravity: Can Google Antigravity help write a LinkedIn retrospective on an agent run?
- GitHub Copilot: Can GitHub Copilot turn a pull request into a LinkedIn story?
- Devin Desktop: Can Devin help write a LinkedIn delivery-impact brief?
- JetBrains Junie: Can JetBrains Junie write a LinkedIn case study from an IDE session?
- ChatGPT: Can ChatGPT write and schedule LinkedIn posts from your analytics?
- Claude: Can Claude turn an interview into an attributed LinkedIn post?
- Manus: Can Manus research a topic and draft a sourced LinkedIn post?
- Perplexity Computer: Can Perplexity Computer prepare a weekly LinkedIn briefing that attributes every item?
- Cue: Can Cue turn your own notes into a LinkedIn draft that keeps your wording?
- Gemini: Can Gemini turn a voice note into a LinkedIn draft saved through Tareno?
- Grok Build: Can Grok Build turn finished engineering work into a LinkedIn case study?
- Kimi Code: Can Kimi Code turn a debugging session into a LinkedIn explanation?
- DeepSeek: Can a DeepSeek-powered client draft a LinkedIn update for an open-source release?
- Paperclip: Can a Paperclip agent draft a manager's monthly retrospective for LinkedIn?
Continue: Connect your account
- Availability
- Custom MCP connection
- Connection
- mcp · Setup documented; client test still required
- Sign-in
- OAuth · Tareno
- Open the client's MCP server settings. Add Tareno with the HTTPS endpoint below using Streamable HTTP. This requires support for authenticated remote MCP, not just local stdio servers.
- Choose OAuth, connect and sign in with your own Tareno account. Check the requested permissions. Refresh the client's tool list after authorization.
- Run list_accounts first. Confirm the returned workspace and account IDs before creating a draft. If no accounts appear, connect a social account in Tareno first.
https://tareno.co/api/mcpUse the same Tareno account across supported clients. Open the relevant setup guide for authentication, first use, reconnect, and revocation.
list_accountsQuestions before you connect it
How does Continue talk to the Tareno MCP server?
Yes, as a remote MCP server using the streamable-http transport that Continue's documentation lists. Tareno has not completed a client test for Continue, and the page does not mention OAuth, so check how your version authenticates.
Where does Continue load MCP servers from?
From a .continue/mcpServers folder in your workspace, where each YAML file defines a server, or from an mcpServers block in the main config. Standalone files need name, version and schema fields at the top.
Why don't I see the Tareno tools in chat?
Continue's documentation says MCP can only be used in agent mode. Switch to agent mode, check the server loaded and ask for `list_accounts` as a read-only test before any draft.
What if my Continue version has no OAuth for remote servers?
Tareno also supports API keys for compatible clients, sent as a bearer header. Keep the key in an environment variable, never in a committed file or a prompt, and replace it in Tareno if it ever leaks.
Will the post publish without my approval?
No. A scheduling request becomes a pending action in Tareno. LinkedIn receives the post only after you approve it in the dashboard, and Continue can only report that it is pending.
Can Continue read our policy from the repository?
Yes, it can read the policy file you point it at, and the post will rest on that document. Ask it to mark anything inferred, and keep the permission inventory out of the public text.