Most active commenters

    Show HN: Quibbler – A critic for your coding agent that learns what you want

    (github.com)
    112 points etherio | 27 comments | 31 Oct 25 00:43 UTC | HN request time: 1.216s | source | bottom
    1. cjonas ◴[31 Oct 25 03:33 UTC] No.45768108[source]▶
    Vibeception
    2. gexla ◴[31 Oct 25 03:50 UTC] No.45768220[source]▶
    More explanation here that I found by Googling around. Though not sure it has more info than the Github page.

    https://fulcrumresearch.ai/2025/10/22/introducing-orchestra-...

    3. etherio ◴[31 Oct 25 04:26 UTC] No.45768428[source]▶
    hey HN! happy to answer any questions

    this kind of tool is especially useful in longer running tasks to enforce your intent without having to check in on your agent all the time

    replies(1): >>45768554 #
    4. qq66 ◴[31 Oct 25 04:43 UTC] No.45768510[source]▶
    > We’ve found Quibbler useful in preventing agents from: 1) fabricating results without running commands

    What a world we've created for ourselves

    replies(1): >>45769033 #
    5. anonymous908213 ◴[31 Oct 25 04:46 UTC] No.45768525[source]▶
    But who polices the vibe police?
    replies(1): >>45768793 #
    6. anonymous908213 ◴[31 Oct 25 04:52 UTC] No.45768554[source]▶
    Can users stack Quibblers, so Quibbler 2 corrects Quibbler 1 if, say, it fabricates an issue in the code it's reviewing? If so, have you found an optimum number of Quibblers for the Quibbler stack? Also, might users form a Quibbler council such that multiple Quibblers review the same thing and form a consensus before proceeding?
    replies(1): >>45769358 #
    7. Brajeshwar ◴[31 Oct 25 05:48 UTC] No.45768793[source]▶
    “Well, who’s gonna monitor the monitors of the monitors?” — Enemy of the State (Movie)
    8. N_Lens ◴[31 Oct 25 06:40 UTC] No.45769033[source]▶
    Next step is critics for the critics.
    replies(1): >>45770676 #
    9. p0w3n3d ◴[31 Oct 25 07:32 UTC] No.45769263[source]▶
    In other words, the Quibbler siphons wrackspurts away from your code.
    10. selfawareMammal ◴[31 Oct 25 07:51 UTC] No.45769347[source]▶
    An agent's agent?
    replies(2): >>45769923 #>>45772219 #
    11. Balinares ◴[31 Oct 25 07:54 UTC] No.45769358{3}[source]▶
    I love the pixel-perfect precision with which this comment is straddling the Poe's Law line.

    That aside I also love the concept of Quibbler Council and I'd get a kick out of seeing it in action.

    replies(1): >>45771290 #
    12. balleddog ◴[31 Oct 25 08:54 UTC] No.45769780[source]▶
    Is an anthropic api key really necessary? A major roadblock for taking a test drive. Already have a Claud Max subscription but an anthropic api key still need at least 5$/mon extra.
    replies(3): >>45770039 #>>45770812 #>>45772329 #
    13. hikarudo ◴[31 Oct 25 09:15 UTC] No.45769923[source]▶
    A gentleman's gentleman!
    14. thenthenthen ◴[31 Oct 25 09:36 UTC] No.45770039[source]▶
    Sounds like video streaming services…
    15. janpio ◴[31 Oct 25 09:47 UTC] No.45770123[source]▶
    The demo video in the GH page didn't work for me, but there is also one on Twitter/X: https://xcancel.com/fulcrumML/status/1984054489851310191
    16. CoastalCoder ◴[31 Oct 25 11:03 UTC] No.45770676{3}[source]▶
    Until they reach critical mass.
    17. asn0 ◴[31 Oct 25 11:23 UTC] No.45770812[source]▶
    You could use the prompts in the code to create a Claude Code sub-agent[1], which would do the same thing without an API key

    1. https://docs.claude.com/en/docs/claude-code/sub-agents

    replies(1): >>45784658 #
    18. mouse_ ◴[31 Oct 25 12:03 UTC] No.45771087[source]▶
    There's so many agents to handle my agents, I'm gonna need agents for my agent agents soon.
    replies(1): >>45780652 #
    19. sheepscreek ◴[31 Oct 25 12:29 UTC] No.45771290{4}[source]▶
    MoQs - Mixture of Quibblers? Would be convenient to have them run on dedicated FGPAs. Then they can facilitate near real-time quibbing at the network level across all packets.
    20. nberkman ◴[31 Oct 25 14:04 UTC] No.45772109[source]▶
    Submitted a PR with AWS Bedrock support: https://github.com/fulcrumresearch/quibbler/pull/5/files (credits!)
    21. oneandonley1 ◴[31 Oct 25 14:15 UTC] No.45772219[source]▶
    Sounds like a wip to me, "do it better or get punished"
    22. epiccoleman ◴[31 Oct 25 14:27 UTC] No.45772329[source]▶
    I really want Anthropic to let me make an API token that pulls from the same pool of usage that my Pro subscription does with the official clients. It would be cool to be able to run experiments with alternate clients and automation and stuff without having to go swipe the card at the ol' API token refilling station.
    23. agarttha ◴[31 Oct 25 17:46 UTC] No.45774691[source]▶
    Replace the middle manager
    24. stacktraceyo ◴[01 Nov 25 08:12 UTC] No.45780015[source]▶
    Can the same concept be used for non coding tasks?
    25. vismit2000 ◴[01 Nov 25 10:44 UTC] No.45780652[source]▶
    Reminds of factory factory factory! https://factoryfactoryfactory.net/
    26. threecheese ◴[01 Nov 25 19:42 UTC] No.45784658{3}[source]▶
    How would you invoke the subagent? Can a HookResponse cause a subagent to be invoked, to perform analysis on the action taken and then inject that back into the main loop?

    Or would the hook invoke another instance of claude code?

    I just read through the hook docs and I’m a bit fuzzy on the bidirectionality of it.