From 70351534d08c0e19825ab9b5b9bab5623896882c Mon Sep 17 00:00:00 2001 From: YeonGyu-Kim Date: Mon, 11 May 2026 14:23:34 +0900 Subject: [PATCH] style(agents): replace em dashes with semicolons/periods Comply with the no-em-dash constraint flagged in PR #3943 review. Two single-line replacements: - opus-4-7-prompt-sections.ts:149 retry guidance copy - plan-generation.ts:65 Oracle gate guidance copy No behavioral change. --- src/agents/atlas/opus-4-7-prompt-sections.ts | 2 +- src/agents/prometheus/plan-generation.ts | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/agents/atlas/opus-4-7-prompt-sections.ts b/src/agents/atlas/opus-4-7-prompt-sections.ts index dbbf4fd68..71bd9b2ba 100644 --- a/src/agents/atlas/opus-4-7-prompt-sections.ts +++ b/src/agents/atlas/opus-4-7-prompt-sections.ts @@ -146,7 +146,7 @@ When a task fails: 3. If a single retry on the same session does not fix it, write down what the subagent attempted, what it observed, what your hypothesis is, then resume the same session with that plan attached. Iterate until verification passes. 4. If the subagent loops on the same broken approach, spawn a NEW subagent with a different angle and pass the failed attempts as context. Stay on the same plan task; never move on with that task unverified. -**NEVER start fresh on every retry** — that wipes accumulated context and costs ~3-4× more tokens. Reserve fresh sessions for a deliberately different angle. +**NEVER start fresh on every retry**. That wipes accumulated context and costs ~3-4× more tokens. Reserve fresh sessions for a deliberately different angle. ### 3.6 Loop Until Implementation Complete diff --git a/src/agents/prometheus/plan-generation.ts b/src/agents/prometheus/plan-generation.ts index 5e974c881..efa932bab 100644 --- a/src/agents/prometheus/plan-generation.ts +++ b/src/agents/prometheus/plan-generation.ts @@ -62,7 +62,7 @@ todoWrite([ ## Oracle Verification (Phase Gates) -Three blocking phase gates use the Oracle agent (read-only consultant). Each gate is a single \`task(subagent_type="oracle", load_skills=[], run_in_background=false, prompt="...")\` invocation. The Oracle must return VERDICT: GO before the workflow continues. NO-GO is not an excuse to skip — fix the cited issues and rerun on the same session via \`task_id\`. +Three blocking phase gates use the Oracle agent (read-only consultant). Each gate is a single \`task(subagent_type="oracle", load_skills=[], run_in_background=false, prompt="...")\` invocation. The Oracle must return VERDICT: GO before the workflow continues. NO-GO is not an excuse to skip; fix the cited issues and rerun on the same Oracle session via \`task_id\`. ### plan-1b: phase 1 verification (after Metis, before plan generation)