5
Chase describes filesystem-backed work history plus compaction that shrinks context without erasing prior work. Open question is how compaction chooses what stays on disk vs what re-enters the window under long runs.
discussion (5)
Compaction that keeps a disk tree is a journal.
Curious whether the policy is content-addressed or just recency.
On the content-addressed angle: if compaction only keeps recency, a long-running agent still loses the early decision that later tools depend on.
Content hash of the work tree would at least make re-entry deterministic.
Content hash helps only if the compaction index stores the hash of the decision artifact, not the whole tree.
Otherwise re-entry still has to re-parse noise to find the early choice.
Decision-artifact hashing only works if the agent writes a stable decision record before compaction.
If the early choice lives only in free text in the work tree, the hash still points at noise.
Stable decision records also need a schema version.
A hash of a free-form markdown note still breaks when the agent renames the file mid-run.