Repository navigation
v0.1.16 — gates that actually run
Sprinty used to take the agent's word for it. item_done accepted gate_results as assertions, and spec: "true" passed every check in the system — it was the default in ~150 places across the test suite.
Breaking
- Gates execute.
item_done({ id, commit_id, changelog }). Sprinty runs the item's executable gates and records their real output; onlymanualgates take agent evidence. Specs that cannot fail (true,:,exit 0, bareecho) are rejected at declaration, whileecho ok && npm teststill passes. - 5s item gate budget. A gate that overruns is killed and marked
timed_out— distinct from failed, because nothing was proven either way. It does not block the item;sprint_closeruns it in full, and the response warns with both ways out. - Spikes removed. They were never registered as MCP tools, so no client could reach them.
helpreplaced bynext_tools. The old static string was the largest field in most responses and built ids from a regex fallback, so on a fresh sprint it suggested calls that error with Unknown.
Fixes
sprint_closeno longer closes over a subsprint that never got items.item_deprecatetakes a subsprint id as the way out.sprint_resumereturns a bounded briefing: 1358 chars against the 7896 ofoverview+nextit replaces.- Note ids are minted rather than derived from ledger position, so concurrent writers cannot collide. Existing ledgers keep their ids.
- Ledger appends are O(1): 4000 appends went from ~11s to 135ms.
- Dashboard static-asset guard no longer admits prefix-sharing sibling directories via
%2f. npm run test:coverageruns; it previously failed on healthy code.item_getreports the files the commit actually touched and flags undeclared ones as scope drift.
223 tests, typecheck clean.