feat: preview and edit shell results before model sharing (#281)

README.md

@@ -94,10 +94,12 @@ A dated, sanitized offline fallback snapshot is available as [HTML](docs/model-c
 Explain the failing test
 !npm test
 /shell-result
-Help me fix it
+/shell-result lines
+/shell-result drop 8-10
+/shell-result send
 ```
 
-`!commands` run locally with **zero model API calls**. Output streams into the console with an exit code. `/shell-result` explicitly shares the latest result with the next model turn; review it for secrets first.
+`!commands` run locally with **zero model API calls**. Output streams into the console with an exit code. `/shell-result` stages and previews the exact bounded attachment, including the command, captured directory, exit status, and any omitted bytes. Edit it with `drop`, `replace`, `mask`, or `redact`, then choose `/shell-result send` or `/shell-result cancel`. Redaction helps identify common secrets but does not guarantee their removal. A later shell command cannot change the staged attachment. For pipes and JSON sessions, `/shell-result send` is the explicit one-step send form.
 
 | Platform | Local shell | Interactive programs |
 |---|---|---|

docs/generated/commands.md

@@ -1,5 +1,5 @@
 <!-- GENERATED FILE: run `npm run docs:generate`; do not edit by hand. -->
-<!-- manifest-digest: sha256:c9524e6cca9e6dc684b59996bcd16fdab2b5a0a0a8007ff3be2b8bff5d7cdc8e -->
+<!-- manifest-digest: sha256:528c55f5f15ba474ef4fb931cea11a1f1275bd261aa7503c08bd9ad09b230e51 -->
 # Generated command reference
 
 This reference is generated from the validated, versioned command manifest. Availability is evaluated at runtime; a listed command may still require authentication, a hosted capability, or local tooling.
@@ -485,7 +485,7 @@ Permission: `unknown` · Availability: `runtime-dependent` · Telemetry: `slash.
 
 #### `/shell-result`
 
-explicitly share the last local shell result with chat \(bounded\)
+preview and edit a bounded local shell result before sharing it
 
 Permission: `network` · Availability: `runtime-dependent` · Telemetry: `slash.shell-result`
 

src/commands/chat.ts

@@ -1354,7 +1354,7 @@ export async function repl(ctx: AppContext, skillOpts: TurnSkillOptions = {}): P
     /** Run one turn without sacrificing an existing type-ahead draft. */
     const runQueuedTurn = async (input: ConsoleInput, authContinuation = false): Promise<"completed" | "aborted" | "failed"> => {
       const sharedShellResult = input.kind === "share";
-      if (input.kind === "share") input = consoleShell.share();
+      if (input.kind === "share") input = consoleShell.share(input);
       if (input.kind === "error") { process.stdout.write(input.message + "\n"); return "completed"; }
       if (input.kind === "empty") return "completed";
       if (input.kind !== "chat") {
@@ -1468,7 +1468,7 @@ export async function repl(ctx: AppContext, skillOpts: TurnSkillOptions = {}): P
         // server's error frame (frame() runs before runTurn throws) — only
         // genuinely unrendered failures (network, fallback-leg errors) need
         // printError's own "✗" line, or the user sees the error twice.
-        const authFailure = await resolveBackend(ctx) === "cloud"
+        const authFailure = !sharedShellResult && await resolveBackend(ctx) === "cloud"
           && authRepair.noteFailure(err, submittedPrompt, turnOutcomeForError(err), receipts);
         if (authFailure && buf.value) heldDraft = buf.value;
         if (authFailure && !ctx.flags.json) {
@@ -1488,7 +1488,7 @@ export async function repl(ctx: AppContext, skillOpts: TurnSkillOptions = {}): P
         // commit() clears the submitted line before the request starts. Put it
         // back only when the user has not typed ahead; otherwise preserve their
         // newer draft and leave the failed submission in history for recall.
-        const recovered = authFailure ? buf.value : recoverSubmittedPrompt(text, buf.value);
+        const recovered = sharedShellResult || authFailure ? buf.value : recoverSubmittedPrompt(text, buf.value);
         if (recovered !== buf.value) {
           buf.clear();
           buf.insert(recovered);
@@ -2037,7 +2037,7 @@ export async function replLines(ctx: AppContext, skillOpts: TurnSkillOptions = {
     if (ConsoleShell.isTerminalCommand(line)) { await consoleShell.terminalCommand(line); continue; }
     let input = classifyConsoleInput(line);
     const sharedShellResult = input.kind === "share";
-    if (input.kind === "share") input = consoleShell.share();
+    if (input.kind === "share") input = consoleShell.share(input, true);
     if (input.kind === "error") { process.stdout.write(input.message + "\n"); if (p) process.stdout.write(p + consoleShell.prompt()); continue; }
     let t = input.kind === "chat" ? input.text : "";
     let authReplay = false;
@@ -2085,7 +2085,7 @@ export async function replLines(ctx: AppContext, skillOpts: TurnSkillOptions = {
       if (p) process.stdout.write(p + consoleShell.prompt());
       continue;
     }
-    if (historyEnabled() && line.trim() !== "/shell-result") appendHistory(line.trim(), historyPath(ctx.flags.cwd));
+    if (historyEnabled() && !sharedShellResult) appendHistory(line.trim(), historyPath(ctx.flags.cwd));
     if (t === "/auth" || t.startsWith("/auth ")) {
       inflight = new AbortController();
       try {
@@ -2166,7 +2166,7 @@ export async function replLines(ctx: AppContext, skillOpts: TurnSkillOptions = {
         const outcome = turnOutcomeForError(err);
         if (ctx.flags.json && outcome) process.stdout.write(turnOutcomeJson(outcome) + "\n");
         else process.stderr.write("\n" + errTheme.dim("✗ canceled — turn discarded") + "\n");
-      } else if (await resolveBackend(ctx) === "cloud" && authRepair.noteFailure(err, submittedPrompt, turnOutcomeForError(err), receipts)) {
+      } else if (!sharedShellResult && await resolveBackend(ctx) === "cloud" && authRepair.noteFailure(err, submittedPrompt, turnOutcomeForError(err), receipts)) {
         if (!ctx.flags.json) process.stderr.write("✗ Hosted credential rejected (401). Task saved. Use /auth login; /auth status shows the credential source.\n");
         printed = true;
       } else if (err instanceof ChatTurnError) {

src/commands/command_manifest_data.ts

@@ -4056,8 +4056,8 @@ export const COMMAND_MANIFEST_SOURCE: readonly CommandManifestEntry[] = [
     "aliases": [],
     "compatibilityAliases": [],
     "deprecatedAliases": [],
-    "summary": "explicitly share the last local shell result with chat (bounded)",
-    "detailedHelp": "/shell-result\nexplicitly share the last local shell result with chat (bounded)",
+    "summary": "preview and edit a bounded local shell result before sharing it",
+    "detailedHelp": "/shell-result previews the exact bounded attachment without contacting a model. Use /shell-result lines to inspect numbered rows; drop <first>[-<last>], replace <line> <text>, mask <literal>, or redact to edit the staged snapshot. /shell-result send shares that snapshot; /shell-result cancel discards it. In pipes or JSON sessions, /shell-result send is the explicit one-step send form. A newer command never replaces a staged preview. Redaction is an aid, not a guarantee.",
     "section": "Steering",
     "hidden": false,
     "permissionClass": "network",
@@ -4084,8 +4084,8 @@ export const COMMAND_MANIFEST_SOURCE: readonly CommandManifestEntry[] = [
       "disposition": "generated"
     },
     "release": {
-      "disposition": "new",
-      "note": "Coding console only; explicit sharing of at most 8 KiB of untrusted local output."
+      "disposition": "changed",
+      "note": "Coding console now previews and edits the command-bound bounded capture before explicit sharing."
     }
   },
   {