A quick clarification on intent, since “Git diff for reconciliation” can sound broader than it is:
VisiGrid CLI is deliberately narrow. It’s not a BI tool, not a data warehouse, and not trying to replace Excel or pandas. It’s a deterministic comparison primitive: given two datasets, it answers what matches, what differs, and what’s missing, with explicit numeric tolerance and machine-checkable output.
The main design constraints were:
same inputs → same outputs (no ambient recalculation)
explicit exit codes so it can run in CI
stable JSON output for downstream automation
runs fully local (no accounts, no network)
I’m pre-v1.0 and very interested in feedback on diff semantics, tolerance handling, or edge cases you’ve hit in reconciliation workflows.
Happy to answer questions or go deep on implementation details.
Hi all — I’m the author.
A quick clarification on intent, since “Git diff for reconciliation” can sound broader than it is:
VisiGrid CLI is deliberately narrow. It’s not a BI tool, not a data warehouse, and not trying to replace Excel or pandas. It’s a deterministic comparison primitive: given two datasets, it answers what matches, what differs, and what’s missing, with explicit numeric tolerance and machine-checkable output.
The main design constraints were:
same inputs → same outputs (no ambient recalculation)
explicit exit codes so it can run in CI
stable JSON output for downstream automation
runs fully local (no accounts, no network)
I’m pre-v1.0 and very interested in feedback on diff semantics, tolerance handling, or edge cases you’ve hit in reconciliation workflows.
Happy to answer questions or go deep on implementation details.