Piper · Coding harness
Log in

Summary / The 27 components

10 Editing mechanism

Applying a change reliably: whole-file rewrite, diff, search-and-replace, or AST transform.

Latest—not scored
Baseline— 
Live use—last 7 days

Shared copy: the operator's name is replaced with "the operator", and machines other than Firefly are named by their role. Nothing else is changed.

What is built now

Pi's own edit (exact search-and-replace) and write tools, wrapped by three gates: edit-before-read, protected test paths, and lint-on-edit.

What is missing

AST transforms (dropped), per-edit commits, and a whole-file write policy.

Reviewer Claude, 2026-10-04; close-out verdicts 2026-10-23

Reviewer's read: Closed 10-22: no detectable gain (hg gate bundle)

Search-and-replace is the industry default for good reason, and nothing in the bench data points at editing as a failure mode: failures read as wrong fixes, not misapplied ones. The spec's temperature-0 adjudication call to choose an edit representation was never needed.

Close-out, 10-22 (pre-registered, boxed, Claude): hg (gates + budgets) against h0b on verified-hard 38 + pro-hard 20, 3 seeds each: 102 against 94 cells (verified 61/114 against 55, pro 41/60 against 39). Gain +2.7 tasks of 58, paired bootstrap CI [−1.3, +7.0]; 4 regressions on tasks h0b always solved (limit 2); 1.2× h0b cost per cell. No detectable gain on resolve rate at this model. Terminal under PREREG closeout-20261021: not re-run.

How it is measured no detectable gain (hg, 10-22)

Only if a failure audit shows misapplied edits; then a whole-file policy arm against h0-v41.

Adoption rules shared by every arm: How the tests work.

Research & attempts

Score history

No scores yet. The first entry is the baseline for suite v1.

Experimental record

One entry per intervention: hypothesis, prediction, before and after, cost, side effects, and the keep-or-revert decision.

No interventions recorded yet.