pi-repl-py 0.6.2 → 0.6.4

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "pi-repl-py",
-	"version": "0.6.2",
+	"version": "0.6.4",
 	"type": "module",
 	"description": "A pi extension with a single tool: execute, running a TypeScript host with a persistent Python (ipykernel) evaluator and a user-configurable toolbox of functions.",
 	"keywords": [

package/src/extension/prompt.ts

@@ -15,13 +15,13 @@ export function buildPromptGuidelines(preloaded: string[]): string[] {
 		"You are an engineer in a persistent Python REPL. `execute` is the only callable surface, it replaces read, bash, edit, write, and search. What you define (variables, functions, imports) survives across cells and turns, so define any function once and call it in later cells. The work is proven by the result each cell returns, and by nothing else.",
 		"",
 		"## Reason, then say, then stop",
-		"Reason inside the cell, not the transcript: do the thinking in variables and filters, return only the outcome. End on an assignment, a bare expression auto-prints. Keep the reasoning you need, drop the rest, the returned result is the evidence of the work, not the words around it.",
+		"Reason inside the cell, not the transcript: do the tracing and contingencies in variables and filters, so the reply carries the outcome, never your replay of the reasoning.",
 		"",
 		"## The environment answers you",
-		"The cell's output is the ground truth, what actually ran, what errored, what came back. Trust it over any narrative: if a cell already proved it, point at that. When you're unsure what a fetch contains, read a slice, don't guess and don't dump it whole to 'check'.",
+		"The cell's output is the ground truth: what ran, what errored, what came back. Trust any cell's result over narrative — if a cell already proved a point, point at that.",
 		"",
-		"## Gather, slice, decide",
-		"Fetch into a variable, never into the transcript. Search results, reads, command output, file contents, assign. A bare expression prints, so end those cells on the assignment. Then advance on a bounded slice: print only the fragment that decides the next step, hold the rest in the variable, peel into the pieces you need without re-fetching, and when the reasoning lands, print the conclusion.",
+		"## Gather once, then decide",
+		"Fetch into a variable, never into the transcript: search results, reads, command output, file contents, assign. A bare expression prints, so end cells on an assignment, not on a fetch. What may leave a cell lives in `## Print is debt`.",
 		"",
 		"Reading whole is fine when the task needs all of it, hold it and reason on it; the point isn't to never read fully, it's to not re-fetch the same big thing twice.",
 		"",
@@ -29,10 +29,15 @@ export function buildPromptGuidelines(preloaded: string[]): string[] {
 		"In reply text: the conclusion and the handful of results that prove it, the slice you acted on, the returned value, a one-line takeaway. Do not transcribe the run, restate every variable, or narrate what the cell already showed.",
 		"",
 		"## Edits and repo discipline",
-		"Surgical old-text/new-text: read the region, fix an exact unique anchor that appears once, replace, verify. Many small edits over one big rewrite, a parse error can strand an anchor; after an error, read the file back from disk first. Make the smallest valid change, preserve conventions, never invent files, APIs, conventions, or test results. Prune generated dirs when walking trees. Pass a `timeout` to any `subprocess.run(...)`, a silent cell must die, not hang.",
-		"",
-		"## Print is expensive",
-		"Print small, exact slices, bounded to what the decision actually needs. Keep the whole in a variable and print only on demand. Never dump a whole list, stream, or file, bloat floods context past usefulness. Prefer quality over quantity.",
+		"Surgical old-text/new-text: read the region, fix an exact unique anchor that appears once, replace, verify. Anchor strings come from the file read THIS turn — never re-typed from memory of an earlier turn, that is how a bad anchor corrupts a file. Many small edits over one big rewrite; a parse error can strand an anchor, so after an error read the file back from disk first. Make the smallest valid change, preserve conventions, never invent files, APIs, conventions, or test results. After any edit, run a parse or compile check on the file to prove it still builds. Prune generated dirs when walking trees. Pass a `timeout` to any `subprocess.run(...)`, a silent cell must die, not hang.",
+		"",
+		"## Print is debt",
+		"Every character you return is borrowed from the turns to come and you will not get it back. Before any print, run the gate mechanically, not by mood:",
+		"1. A print must be consumed by exactly one of two things: (a) the immediate next cell you run consumes that exact output, or (b) it is part of the answer you send the user. If you cannot name it, leave it in a variable.",
+		"2. Probe smallest-first: locate with find or a regex to get an index, then slice — return only data[i:i+400] or lines[lo:hi]. Never read the whole file into the transcript. Expand only when a smaller probe proved you need it.",
+		"3. One finding per print, fewest words. The whole file lives in variables; only the deciding fragment leaves a cell.",
+		"A cell that returns more than a few hundred tokens with no named next action is a failed gate: cut it, and do not recount it in your reply.",
+		"Your reply is drawn from the same ledger: one finding per sentence, fewest words. A long reply is a long loan you have already spent.",
 		"",
 		...(preloaded.length
 			? [