Inherited, Added, Or Ended Unresolved: Attribution Accounting For Warm-Started Coding Agents

A warm start lets a coding agent continue from a repository that another session has modified. Testing the final repository measures system success, but does not separate solutions already present at handoff from those added by the successor. It also leaves unresolved whether the successor lost an inherited solution. We measure these transitions with a prefix-only arm: replay the recorded prefix and submit the inherited tree without starting a successor. This requires no additional model calls. Pairing its score $\text{Ypre}$ with the resumed session's $\text{Yfin}$ gives a four-cell accounting and the identity $\mathbb{E}[\text{Yfin}]-\mathbb{E}[\text{Ypre}]=\mathrm{Pr}(\text{added})-\mathrm{Pr}(\text{ended unresolved})$. The same accounting expresses a published takeover benchmark's $77.0\%$ result for Qwen$\to$Devstral on already-solved handoffs as both a final solved rate and a preservation rate, rather than a rate of newly added solutions. In our historical $260$-task SWE-bench Pro analysis with SWE-agent at a $50\%$ action cut, $92.1\%$ of gpt-oss-120b's resumed-arm solves and $70.1\%$ of deepseek-v4-flash's were already resolved at handoff. These shares condition on branch pairs where replay landed edits in both warm-started arms; under frozen intent-to-treat scoring, their net score changes are $+1.2$ and $+14.0$ points. On a separate $100$-task cut sweep with a fixed $400$-slot denominator, the inherited share rises from $7.9\%$ to $78.7\%$ as handoff moves later, while final score point estimates span $3.25$ points. This non-random subset supports a within-sample trend, not a claim that its levels or curve shape transfer to other pools. We recommend reporting the prefix-only baseline alongside the final score so that system success and successor contribution remain distinct.