ccd-backlog
id = 173
kind = "item"
section = "architecture"
status = "ready"
effort = "large"
impact = "medium"
priority_rank = 270
depends_on = []
Context
Pituitary currently operates on a single repo. But many teams manage specs across multiple repos — a platform repo with core specs, service repos with implementation specs, a docs repo with guides.
Questions this RFC should answer
- How do specs in repo A reference specs in repo B?
- How does drift detection work across repo boundaries?
- Is this a multi-index query or an index-merging strategy?
- How does the extension architecture (RFC 0002) support cross-repo sources?
- What's the config model? (multiple workspace roots? federated index?)
Why now
The positioning explicitly targets "development teams for multi-repo codebases and bigger projects." This is the scaling story beyond single-repo. Lower priority than Intent Graph and Confidence Calibration since it's a larger architectural investment, but should be designed before the first enterprise adopters hit the wall.
Context
Pituitary currently operates on a single repo. But many teams manage specs across multiple repos — a platform repo with core specs, service repos with implementation specs, a docs repo with guides.
Questions this RFC should answer
Why now
The positioning explicitly targets "development teams for multi-repo codebases and bigger projects." This is the scaling story beyond single-repo. Lower priority than Intent Graph and Confidence Calibration since it's a larger architectural investment, but should be designed before the first enterprise adopters hit the wall.