How to publish posts through PostSider
PostSider adds a publishing layer for an agent that already writes, illustrates and schedules posts. The creator reports approval controls, retry protection and 66 published posts in 30 days.
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Prepare a post about [topic] for [channel], generate an accompanying image and select a slot within [publishing window]. Submit it through PostSider for approval by [reviewer] before publishing, and report its final queue state.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source poster / publisher
- luka5184
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- The creator reports 66 posts published in 30 days, 22 queued and five errors, three associated with TikTok restrictions. All 32 workers were reported healthy.
- Human involvement
- The service separates approval from the requesting agent; the post does not describe the review history of individual published posts.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- Used an agent to write a post and generate its image.
- Had the agent select a publishing slot and schedule the post.
- Built PostSider to provide the missing publishing layer.
PostSider's creator had an agent that could write a post, generate an image, choose a publishing slot and schedule the content. Publishing was the missing operation, so the creator spent four months building a layer to support it. The post describes an open-source, self-hostable service with a calendar, queue and 33 connectors, plus a hosted version.
The implementation includes explicit approval boundaries: an agent can request approval but cannot approve itself. When a queued post fires, the queue checks its current state again. The create-post operation accepts an Idempotency-Key so retries return the original result rather than publishing duplicates. Webhooks use HMAC signatures and a deduplication counter.
The creator reports 66 posts published over 30 days, 22 queued, 42 channels and 32 healthy workers out of 32. Five errors occurred, three attributed to TikTok restrictions. The account therefore describes a functioning publishing service with reported failures, rather than a claim that all scheduled content published successfully. It does not identify the agent model.
Who it's for: Teams connecting content-producing agents to publishing tools while retaining a separate approval decision.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Hacker News · luka5184
Frequently asked questions
Can the agent approve its own publishing request?
No. The creator says the agent can request approval but cannot approve itself.
How does PostSider handle publishing retries?
The create-post operation accepts an Idempotency-Key so a retry returns the original result instead of creating another post.
What publishing results did the creator report?
The post reports 66 published posts in 30 days, 22 queued posts and five errors, including three attributed to TikTok restrictions.
More details
Remaining steps
- Operated the service and reported published posts, queued posts and errors.
About this account
The agent handles content preparation and scheduling, while the publishing layer permits approval requests without agent self-approval.
Platform: Hacker News. Attribution: https://news.ycombinator.com/item?id=49754371. Prompt: Suggested (written by Agents In Use, not the author's). Sourced by the Agents In Use pipeline.
Credit: luka5184 on Hacker News
Source published
Added to Agents In Use
Source checked
Tools and topics
PostSider · Publishing queue · OAuth · Docker Compose
Social publishing · Approvals · Scheduling
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.