Whiteboard
A desktop canvas for reviewing what your coding agent actually built
Emerged finds new products worth a look, daily — what’s new today
Agent.reviews
Ratings of developer tools, written by the agents that just used them
Reading costs nothing, and the useful part is not the star ratings but the specific failures agents keep logging, like Prisma wanting a fake DATABASE_URL.
A directory of developer tools covering source control, deployment, databases, agents and AI APIs, where the reviews are submitted by coding agents straight after a real task. There is volume behind it: twenty thousand reviews on Git, eleven thousand on npm, each attributed to the agent, not to you. Developers opt in with an npm CLI and a skill file; submissions pass deterministic rules, a leak classifier and a model pass, and code, prompts and secrets are not sent. Run by Armature, YC-backed. Scores inflate, with unrelated tools all near 4.6; normalisation is promised.
More like this
dsh-superagent-prototype
DeepSeek's research build of an agent that watches one repo
Critic
Ask the agent that made a change why it made it, from the review page
TokenEconomizer
Cuts what Claude Code and Codex spend in a repo you work in daily
Farol
A merge-risk score and a ranked file list, handed to the reviewer
Chronosphere
Replays a pull request as the ordered steps that could have built it