Loading video...

Video Failed to Load

Go Home

Introducing watchmen. It writes the skill files your coding agents should already have from your own sessions, shared across Claude Code, Codex, and pi, so you stop paying tokens to re-explain what your agent learned last week. Local and open-source.

44,203 views • 6 months ago •via X (Twitter)

19 Comments

Dria's profile picture
Dria6 months ago

Coding agents do carry context now, but it's siloed per tool and you maintain it by hand. CLAUDE.md doesn't help Codex and AGENTS.md doesn't help Claude Code, so both are only as good as what you wrote down. Watchmen mines what you did and writes both from one corpus.

Dria's profile picture
Dria6 months ago

It runs locally and in the background, behind every agent you already use, watching what you actually do and writing the recurring work to disk as runnable skill bundles and a workspace brief. You install it once and never change how you work.

Dria's profile picture
Dria6 months ago

Watchmen measures its own effect with a per-project impact card that tracks median tool errors, prompts, and cost per session before and after your skills land, over 16 weeks. The card is labeled correlation only, not a controlled experiment, so we don't oversell the signal.

Dria's profile picture
Dria6 months ago

Watchmen is local by default, MIT licensed, and runs on macOS and Linux. Your transcripts never leave your machine except for the LLM calls you opted into, and it's available now at:

Alex's profile picture
Alex6 months ago

I find auto-skill creation is a quick method to drown in slop. Creating skills is just part of the equation. There needs to be an eval harness to review which skills actually get used, which ones sit there with their descriptions, eating tokens in your context window. And which ones are used but don't improve the outputs... Most of our agents aren't really suffering from a lack of agent skills. On the contrary, they're already drowning in noise.

Timur Yessenov's profile picture
Timur Yessenov6 months ago

I like the skill-file angle, but I’d keep one rule strict: learned patterns need provenance. Which session produced it, what failure it fixed, and when to delete it.

Brian Johnson's profile picture
Brian Johnson6 months ago

This is the right direction. Re-teaching agents the same repo lore every week is quiet token burn and quiet workflow debt.

Prismor's profile picture
Prismor6 months ago

This is a great idea! Claude have gotten slower lately and expensive

49Agents IDE - for 10x agentic coders's profile picture
49Agents IDE - for 10x agentic coders6 months ago

local and open source is the key here. paying tokens to re-explain what your agent learned last week adds up fast, especially on long projects. making that shared across codex, claude code, and pi is a solid value proposition

AGentIk's profile picture
AGentIk6 months ago

Nice ! Interesting idea

Vik's profile picture
Vik6 months ago

Skill files are underrated because they turn “explain it again” into infrastructure. That’s where agents start compounding.

Joe Fernandez's profile picture
Joe Fernandez6 months ago

how do you handle the sheer amount of skills that can created from something like this? I built a manual version of this and overtime there were just too many - they didn't fit in the budget allotted for skill descriptions in CC/Codex.

paul nescau's profile picture
paul nescau6 months ago

potential to get blasted with useless skills seems high, specially because you are not always asking for the same thing across different repos... you should demonstrate how it works beyond this simple video, together with the dashboard.

hacsceo's profile picture
hacsceo6 months ago

How it works. Technical teardown. Interesting project!

MartinM's profile picture
MartinM6 months ago

looks cool, would prefer to use my existing openai subscription or api key instead of openrouter though

Qasim Khan's profile picture
Qasim Khan6 months ago

Does it work on windows?

dalu's profile picture
dalu6 months ago

is it vibe coded?

ClaudeAPI's profile picture
ClaudeAPI5 months ago

hi , we want to cooperate with you as a inner PR, is there any ways i can connect you so we discuss some details?

jestermolecule's profile picture
jestermolecule6 months ago

this is so cool dria! could i make a pսmpfun for watchmen and make a donation directly to your github for this?

Related Videos