docs(sf): align agent escalation guidance with already-resolved reality

The execute-task escalation guidance claimed the user "can review or
override later via /sf escalate". Commit c1ce9aac1 already made the
already-resolved message explicit that auto-accepted decisions can't
be retroactively undone — the carry-forward into downstream tasks
happens before any operator could intervene.

Updated the agent-facing guidance to match: auto-mode accepts +
persists as memory + carries forward; the operator gets the audit
trail via /sf escalate list --all but the executed work stands. This
shifts the agent's incentive toward thorough rationale capture (since
that's what survives) rather than the false comfort of "the user can
fix it later".

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Mikael Hugo 2026-05-02 23:43:18 +02:00
parent 5cc2522646
commit fec6c293bf

View file

@ -2523,7 +2523,7 @@ export async function buildExecuteTaskPrompt(
"}",
"```",
"",
"Provide 24 options with concrete tradeoffs. The recommendation must reference one of the option ids. Auto-mode accepts your recommendation and carries it forward as a hard constraint for downstream tasks; the user can review or override later via `/sf escalate`. Set `continueWithDefault: false` only when the choice is severe enough that the loop should pause for human review even in auto-mode (rare).",
"Provide 24 options with concrete tradeoffs. The recommendation must reference one of the option ids. Auto-mode accepts your recommendation, persists the choice + rationale as a memory, and carries it forward as a hard constraint for downstream tasks. The operator can review the audit trail later via `/sf escalate list --all`; the executed work itself can't be retroactively undone, so document your reasoning thoroughly. Set `continueWithDefault: false` only when the choice is severe enough that the loop should pause for human review even in auto-mode (rare).",
].join("\n")
: "";