You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A session usually starts with one clear goal in my head. But then it runs for a while — multiple PRs, several rounds of code review, a few tangents — and by the time I come back to it, the original reason that session exists has been buried under everything that happened since.
At that point my only option is to ask the LLM "what were we doing here again?" and hope the answer is still in context. That's a weirdly indirect way to recover something I knew perfectly well at the start.
The request
Let me attach a note to a chat/session. Free-form text, written by me, that stays put and isn't affected by compaction or context loss.
Then surface it on hover. The hover card already shows the useful stuff — repo, machine, model. If a note exists, show it there too:
my-repo · feature/billing-webhooks
MacBook Pro · Opus 5
Note: Stripe webhook retries are dropping events —
this session is the fix + backfill script. Don't merge
until the backfill has been dry-run against staging.
Now I can scan the sidebar and instantly know what each session is for, without opening any of them.
Why a manual note instead of an auto-summary
An auto-generated summary tells me what the session has been doing. A note tells me what it's for — the intent, the constraint, the thing I told myself not to forget. Those are different, and the second one only I can write. Auto-summarizing would be a nice complement, but it doesn't replace this.
Nice-to-haves (not required)
Editable from the sidebar's right-click menu, not buried in settings
A small indicator on sessions that have a note, so I know a hover is worth it
Notes searchable alongside thread titles
Persist per session so it survives restarts
Even the plain version — a text field plus hover display — would solve the whole problem for me.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
The problem
A session usually starts with one clear goal in my head. But then it runs for a while — multiple PRs, several rounds of code review, a few tangents — and by the time I come back to it, the original reason that session exists has been buried under everything that happened since.
At that point my only option is to ask the LLM "what were we doing here again?" and hope the answer is still in context. That's a weirdly indirect way to recover something I knew perfectly well at the start.
The request
Let me attach a note to a chat/session. Free-form text, written by me, that stays put and isn't affected by compaction or context loss.
Then surface it on hover. The hover card already shows the useful stuff — repo, machine, model. If a note exists, show it there too:
Now I can scan the sidebar and instantly know what each session is for, without opening any of them.
Why a manual note instead of an auto-summary
An auto-generated summary tells me what the session has been doing. A note tells me what it's for — the intent, the constraint, the thing I told myself not to forget. Those are different, and the second one only I can write. Auto-summarizing would be a nice complement, but it doesn't replace this.
Nice-to-haves (not required)
Even the plain version — a text field plus hover display — would solve the whole problem for me.
All reactions