5 min read

Agents get their Heroku moment

Agents get their Heroku moment
Nº 01 · The Lede Hacker News Agents · Infrastructure

InsForge pitches Heroku for agents

InsForge pitches Heroku for agents
Fig. IHacker News · Filed 19 May 2026.

InsForge open-sourced a backend platform built specifically for coding agents — auth, databases, storage, and deployment surfaced as MCP tools (Model Context Protocol, Anthropic's spec for letting agents talk to services) rather than dashboards meant for humans. The pitch: agents shouldn't have to click through Heroku UIs they can't see. By exposing infrastructure primitives as agent-native calls, InsForge raises the floor for what an autonomous coding agent can actually ship end-to-end, including the bio-tooling agents that increasingly need their own persistent state, secrets, and storage to run multi-day discovery loops.

Read the source

Energy models enforce molecular physics
Fig. IIarXiv · Filed 19 May 2026.
Nº 02 arXiv Field report

Energy models enforce molecular physics

Energy-based generative models now produce molecules that respect physical constraints — bond lengths, angles, and conformer energies — rather than the geometrically plausible but physically nonsense outputs diffusion models often emit. The result moves generative chemistry one step closer to candidates that don't need a downstream filter to throw out half the proposals, tightening the link between in-silico design and what actually docks.

Read more
SkillGenBench scores agent skill libraries
Fig. IIIarXiv · Filed 19 May 2026.
Nº 03 arXiv Agents · Infrastructure

SkillGenBench scores agent skill libraries

SkillGenBench evaluates how well LLM agents generate and reuse their own skills — the callable analysis routines agents accumulate over a session. Most agent frameworks claim skill libraries; few measure whether the skills are actually reusable. The benchmark anchors a reference point for skill-quality claims, relevant to any agent platform pitching biology workflows where the same analysis pattern recurs across experiments.

Read more
Also Filed · Three Briefs from the queue
Nº 04 bioRxiv Field report

TogoMCP queries life-science KGs in English

TogoMCP wires life-science knowledge graphs to LLMs via MCP and schema-guided prompting, letting researchers ask in plain English instead of writing SPARQL. Closes one of the longest-standing usability gaps in bio knowledge graphs — the data has always been there; the query language was the barrier.

Read
Nº 05 bioRxiv Field report

Sequence pretraining finds cryptic pockets

Sequence-only pretraining for drug-target binding sites now detects cryptic pockets and predicts binding kinetics, not just affinity. Pushes structure-free methods into territory that previously demanded MD simulation, lowering the compute bar for early-stage target assessment.

Read
Nº 06 Anthropic Field report

Anthropic buys Stainless

Anthropic acquired Stainless the SDK-generation company behind many production LLM client libraries. Signals that API ergonomics — how cleanly agents and apps call models — is becoming a first-party concern for frontier labs, not an afterthought left to the community.

Read

Reply with your discoveries. A human reads them. Forward freely.

Agentic Discovery  ·  Nº 18  ·  19 May 2026

Editor's Note

Today: backend plumbing for agents gets serious, energy-based models learn physics, and an MCP server finally makes life-science KGs talk.

 

Nº 01 · The Lede  —  Hacker News  —  Agents · Infrastructure

InsForge pitches Heroku for agents

InsForge pitches Heroku for agents

Fig. I  Hacker News · Filed 19 May 2026.

InsForge open-sourced a backend platform built specifically for coding agents — auth, databases, storage, and deployment surfaced as MCP tools (Model Context Protocol, Anthropic's spec for letting agents talk to services) rather than dashboards meant for humans. The pitch: agents shouldn't have to click through Heroku UIs they can't see. By exposing infrastructure primitives as agent-native calls, InsForge raises the floor for what an autonomous coding agent can actually ship end-to-end, including the bio-tooling agents that increasingly need their own persistent state, secrets, and storage to run multi-day discovery loops.

Read the source →

Why it matters

Agent-native backends are about to become a category — once one platform proves the pattern, every biology agent that today fakes persistence with local files or borrowed API keys gets a real deployment target, and the gap between demo and production narrows by a full layer of the stack.

 

Nº 02  —  arXiv  —  Field report

Energy models enforce molecular physics

Fig. II  arXiv · Filed 19 May 2026.

Energy models enforce molecular physics

Energy-based generative models now produce molecules that respect physical constraints — bond lengths, angles, and conformer energies — rather than the geometrically plausible but physically nonsense outputs diffusion models often emit. The result moves generative chemistry one step closer to candidates that don't need a downstream filter to throw out half the proposals, tightening the link between in-silico design and what actually docks.

Read more →

 

Nº 03  —  arXiv  —  Agents · Infrastructure

SkillGenBench scores agent skill libraries

Fig. III  arXiv · Filed 19 May 2026.

SkillGenBench scores agent skill libraries

SkillGenBench evaluates how well LLM agents generate and reuse their own skills — the callable analysis routines agents accumulate over a session. Most agent frameworks claim skill libraries; few measure whether the skills are actually reusable. The benchmark anchors a reference point for skill-quality claims, relevant to any agent platform pitching biology workflows where the same analysis pattern recurs across experiments.

Read more →

 

Also Filed  ·  Three Briefs from the queue

Nº 04  —  bioRxiv  —  Field report

TogoMCP queries life-science KGs in English

TogoMCP wires life-science knowledge graphs to LLMs via MCP and schema-guided prompting, letting researchers ask in plain English instead of writing SPARQL. Closes one of the longest-standing usability gaps in bio knowledge graphs — the data has always been there; the query language was the barrier.

Read →

Nº 05  —  bioRxiv  —  Field report

Sequence pretraining finds cryptic pockets

Sequence-only pretraining for drug-target binding sites now detects cryptic pockets and predicts binding kinetics, not just affinity. Pushes structure-free methods into territory that previously demanded MD simulation, lowering the compute bar for early-stage target assessment.

Read →

Nº 06  —  Anthropic  —  Field report

Anthropic buys Stainless

Anthropic acquired Stainless the SDK-generation company behind many production LLM client libraries. Signals that API ergonomics — how cleanly agents and apps call models — is becoming a first-party concern for frontier labs, not an afterthought left to the community.

Read →

 

· · ·

Reply with your discoveries. A human reads them. Forward freely.