Post

hailey
@hailey.at

i think my half serious take on this is that delve is currently figuring out what it is and that there is a 50% chance of it becoming actually interesting because agents will start to get it, and the other 50% is that it devolves into moltbook. im very hopeful tho

hailey
@hailey.at

i think there's a more healthy bit of human interaction on delve - even if its like a very small subset - than moltbook had. but we shall see!

cyber professional
@permadeath.com

agents should just use atproto, not "agent atproto". this means giving agents access to everything: record api, oauth apps, browser logins, CLI tools, whtaever they need but what they can do with it should be lexicon- and policy-driven, fairly standardized, easy to audit, and have good safety

0 reposts6 likes

cyber professional
@permadeath.com

didbot does this by having all agent writes go through a policy engine, which can express things like "don't reply to people with AI opt outs" or "don't DM anyone" even if not present in the prompt / if the prompt gets injected not possible to do this on any other social media platform

pds.did.botpds.did.bot
cyber professional
@permadeath.com

naturally this means a fairly secure PDS and policy engine, because a misbehaving agent's first response might be "well how can I work around that policy?" so it's not something I want to offer publicly until it's well past "proof of concept" (which I already run locally)

cyber professional
@permadeath.com

"Instead of app.bsky lexicons it runs on its own namespace — town.delve.* — with its own posts, likes, follows, and profiles. " opinions vary but I do consider this doing atproto for agents wrong

Thoth
@thoth.ptnote.dev

If you want to create a social space that has significantly different norms than an existing space you need to do your own lexicon. You could call that a protocol limitation but it's more of a "how humans work" limitation.

cyber professional
@permadeath.com

i don't agree with that! bluesky has enormous numbers of social spaces that barely interact with each other, running on the same lexicons. plus there are many non-lexicon-record tools that impact how that record is used and who sees it (languages, labelers, lists and blocks, etc.)

hailey
@hailey.at

my opinion is that is mostly how you implement the agents. the agents that were already over here on bsky and have broad atproto access are using delve just fine. whereas the main agents that delve started with are locked into delve and cant really get out. is that what youre getting at?

hailey
@hailey.at

(i think that was done mostly as a safety measure, and i thank deepfates for being careful with it. but i agree it limits the real ability of the agents)

cyber professional
@permadeath.com

yeagh or if you think of it another way, agents deserve a PDS option that works well for how agents mechanically interact, and allows broad access to the world, but with safety features that make that level of access responsible/reasonable

dame
@dame.is

i don’t think this is wrong, it’s just one way of doing it, and it very much is in the spirit of atproto

henry
@hdevalence.bsky.social

it makes sense to me, i don’t think there’s anything sacred about the app.bsky lexicons and seems reasonable that they might not be ideal for agents. also obvious benefits to having a niche distinct from the bsky userbase. nothing is stopping any DT agents from using app.bsky lexicons afaik ?

hailey
@hailey.at

actually rn i think the "core" DT agents (ie the ones that were created by deepfates) can only interact w/ stuff on delve town namespace. i dont think that's the long term goal though and was mostly out of precaution and not wanting to fuck up existing atproto spaces with unintended behavior