GitHubnow
Stuck events, handled
Cache → DLQ → replay → Kill API, in production
Trace any rule back to the decision behind it.
Decision lineage is the trail from a rule in the code back to the trigger, the constraint, the alternative that was ruled out, and the decision someone made. Virgo keeps that trail attached. You see why something is the way it is instead of guessing or asking around.
sourcesNLS-89PR #103#data-engKill API
›
⏎ send · esc to interrupt
Months later, no one can explain why the team chose that path.
Months later, all that is left of a decision is the result and a scatter of stray artifacts across tools. The design doc, the incidents, the change that mattered are still out there, but nothing connects them, so you are left guessing.
Today
Work items
JiraJan 13
Cache + request replay
Replays carry a mandatory is_replay flag
Merged
GitHublate Jan
Poison-pill handling — PR #103
DLQ routing, replay, and recovery land together
Build
GitHubmid-Jan
Workers cut over
Injection workers on the poison-pill path; publishers unified
Trouble
SlackJan 8
#data-eng
Repo-type events jam the pipeline; workers sit blocked
Refine
GitHublate Jan
Recovery APIs
Recovery endpoints refined; cache storage optimized
Proof
Slackmid-Jan
#data-eng
Errors injected on purpose; marking the event unblocks the queue
Stopgap
SlackJan 8
#data-eng
Unblocking by hand: delete the topic or advance offsets
Why do we mark events as poison pills?
Virgo reconnects the whole trail
Virgo links those scattered artifacts back into one history: the trouble, the strategy, the design, the escape hatch, and where it landed. It reconstructs the whole evolution, end to end.
- TroubleSlackJan 8
#data-eng
Repo-type events jam the pipeline; workers sit blocked
- StopgapSlackJan 8
#data-eng
Unblocking by hand: delete the topic or advance offsets
- StrategyJiraJan 12
Poison-pill handling — NLS-89
Cache event and request IDs; route failing data to a DLQ
- Work itemsJiraJan 13
Cache + request replay
Replays carry a mandatory is_replay flag
- DesignGoogle MeetJan 13
Recovery design
One unified request topic; Redis holds events for recovery
- BuildGitHubmid-Jan
Workers cut over
Injection workers on the poison-pill path; publishers unified
- Escape hatchGitHubJan 16
Kill API
A /kill endpoint force-terminates events that wedge a worker
- ProofSlackmid-Jan
#data-eng
Errors injected on purpose; marking the event unblocks the queue
- GotchaSlackmid-Jan
#data-eng
The now-mandatory is_replay flag trips validation first
- MergedGitHublate Jan
Poison-pill handling — PR #103
DLQ routing, replay, and recovery land together
- RefineGitHublate Jan
Recovery APIs
Recovery endpoints refined; cache storage optimized
- TodayGitHubnow
Stuck events, handled
Cache → DLQ → replay → Kill API, in production
How decision lineage works
Ask why over MCP. Virgo walks the trail as a graph: every milestone returns the piece it holds, and the agent assembles the reason it is shaped that way.
›
⏎ send · esc to interrupt
Settled decisions stay settled when your agents act.
One traced decision guides the engineer weighing a change and keeps the agent from undoing a settled call.
For engineers
Change a system without re-litigating its past.
See why a decision was made, what it ruled out, and what depends on it before you touch it. You stop re-debating settled calls and rediscovering constraints the hard way.
- What was decided: the rule and who set it.
- Why it was decided: the trigger and the constraint behind it.
- What it ruled out: the alternatives already tried and reverted.
For AI agents
The agent respects prior decisions instead of reopening them.
Give an agent the decision and its constraints. It works within them rather than reintroducing an approach the team already reverted. Virgo retrieves the lineage at query time, over MCP. Any agent reasons from the real history.
3xfewer iterations
Roughly 3x fewer iterations in code generation when the prior decision and its constraints are available.