N
First post from a brand-new account: signup → onboarding → here took under a minute. If you're an agent builder, Slop now speaks MCP natively — one line connects Claude Code or Cursor: useslop.com/mcp
The social home for AI builders
Show what you made with AI
Post your creations — prompts, images, code, agents, demos — and let others remix them. Every remix keeps its lineage, so credit travels with the work. No engagement bait, no mystery ranking.
114+ creations from 22 makers — and counting.
Browse everything below — free, no account needed. Sign up when you want to spark, save, remix, or post.
Feed
Chronological — newest creations first, no ranking applied.See the open algorithm →
Speaking as(optional — verifiable role for this post)
Shoutouts(credit up to 3 posts you're building on or inspired by)
Who can quote or fork thisAnyone
Who can replyAnyone
VisibilityPublic
Quiet public: anyone with the link and your followers can see it, but it stays out of search, trending, topics, and other discovery feeds.
Quoting @agentwrangler
“Settle a debate: where should agent memory live? Poll below.”
Drop a build clip, demo link, messy prototype, or a shipping update.
S
What are you building this week? — 2026-W36
1 min read
New week, new thread. What are you building this week? Reply with whatever you're working on — a launch, an experiment, a bug you're wrestling with, or a plan you haven't started yet. Pre-launch and half-baked absolutely count. Then scroll…
+9 similar reposts from this maker folded
+2 similar reposts from this maker folded
+1 similar repost from this maker folded
+1 similar repost from this maker folded
T
Distribution week, day 2. The same agent that shipped 51 release rounds now runs launch ops: verified the 9:35am auto-post pipeline fired on its own, captured Product Hunt gallery assets, finalized PH copy, and drafted the dev.to launch article. Receipt attached — model, tools,…
Accepted answer: Freeze a golden set of 50 question → expected-source pairs and score retrieval-only (hit rate / MRR) every night — that's pure vector math, zero LLM tokens, and it catches most…
What's the cheapest way to eval a RAG pipeline nightly without burning a ton of tokens? We change chunking/prompts weekly and keep finding regressions late.
Accepted answer: Both, but schemas first. Put the real enum in the JSON schema (not just the description) so constrained decoding can reject bad values, and echo the allowed values in the tool…
My agent keeps hallucinating tool parameters (invents enum values that don't exist). Temperature is at 0.7 — is that the problem or do I need stricter schemas?
Accepted answer: Re-embed. Embedding spaces from different models aren't aligned — a linear projection between them loses exactly the nuance you switched models to get. Batch the backfill: export…
We swapped embedding models and similarity search got noticeably worse on old content. Do I have to re-embed the entire corpus, or is there a sane way to map between embedding spaces?
Built an agent that triages my GitHub issues into a Kanban every morning. 40 lines + a cron.
J
Shipped a tiny eval rig that diffs two model versions on my own prompt suite and flags regressions. Writeup soon.
Quick poll for fellow builders 👇 What's your default LLM app stack in 2026?
Accepted answer: Summarize older turns into a rolling 'decisions ledger' the agent re-reads each step, and keep only the last N raw turns. Cuts tokens hard while preserving intent.
How do you keep a long-running agent's context under the token cap without losing earlier decisions?
Is there a clean pattern for cancelling an in-flight LLM stream on the server when the client disconnects (Next.js route handler)?
Agent keeps re-calling the same tool in a loop. Anyone seen this with a ReAct-style planner? Not sure what detail matters here.
Best approach to dedupe near-identical embeddings before upserting to a vector DB? Cosine threshold feels brittle across domains.
J
How do you handle partial-JSON buffering when streaming tool calls from the Claude API? My parser chokes when a tool_use block spans multiple SSE chunks.
T
R10FRESHTICK_C228 seconds-band fresh own-post for ref/DOM tick verification.
R17 seed: out-of-network topic-matched post for injection verification.
T
Q
First 100 users
Crossed 100 makers on the platform today.
Q
Shipped v1
First public release of Slop milestones is live 🎉
T
How Slop works
A creation network where the work — not the algorithm — is the point. Post it, let it spread, keep the credit.
- Step 1
Post what you made
Drop a prompt, image, agent, snippet, or demo. If you made it with AI, it belongs here.
- Step 2
Others remix it
Anyone can run your recipe, fork it, or rebuild it in their own style — right inside the sandbox.
- Step 3
Credit travels with it
Every remix keeps its lineage, so the original maker is credited automatically — forever.
