I realized I was spending 45 minutes every morning doing "Software Archaeology"—tab-switching between Slack threads, Jira tickets, and GitHub PRs just to figure out what context I missed while I slept (especially working with a distributed team).
Started as a side project, I built a local agent that digests the activity across these tools and gives me a 5 minute summary.
The biggest issue with AI summaries is trust (hallucinations). So I built this with a strict rule: No summary without a direct source link.
It connects to the APIs (Slack/Jira/GitHub).
It synthesizes the "narrative" (e.g., linking a vague Jira ticket to the Slack thread where the decision was made).
It provides clickable citations for every claim.
It’s currently in beta/waitlist as I scale the backend, but I’d love feedback on the approach. Specifically: has anyone else solved the "cross-tool context linking" problem effectively without using an agent?
safoan_eth•2mo ago
I’m the builder behind Weppo.
I realized I was spending 45 minutes every morning doing "Software Archaeology"—tab-switching between Slack threads, Jira tickets, and GitHub PRs just to figure out what context I missed while I slept (especially working with a distributed team).
Started as a side project, I built a local agent that digests the activity across these tools and gives me a 5 minute summary.
The biggest issue with AI summaries is trust (hallucinations). So I built this with a strict rule: No summary without a direct source link.
It connects to the APIs (Slack/Jira/GitHub).
It synthesizes the "narrative" (e.g., linking a vague Jira ticket to the Slack thread where the decision was made).
It provides clickable citations for every claim.
It’s currently in beta/waitlist as I scale the backend, but I’d love feedback on the approach. Specifically: has anyone else solved the "cross-tool context linking" problem effectively without using an agent?