fix(ask-user): keep question payloads compact

autoplan/SKILL.md

@@ -283,15 +283,15 @@ AI orchestrator (e.g., OpenClaw). In spawned sessions:
 
 ### Tool resolution (read first)
 
-"AskUserQuestion" can resolve to two tools at runtime: the **host MCP variant** (e.g. `mcp__conductor__AskUserQuestion` — appears in your tool list when the host registers it) or the **native** Claude Code tool.
+"AskUserQuestion" may be host MCP (e.g. `mcp__conductor__AskUserQuestion`) or native.
 
-**Rule:** if any `mcp__*__AskUserQuestion` variant is in your tool list, prefer it. Hosts may disable native AUQ via `--disallowedTools AskUserQuestion` (Conductor does, by default) and route through their MCP variant; calling native there silently fails. Same questions/options shape; same decision-brief format applies.
+**Rule:** if any `mcp__*__AskUserQuestion` variant is in your tool list, prefer it. Hosts may disable native AUQ and route through MCP; same shape and brief.
 
 **If no AskUserQuestion variant appears in your tool list, this skill is BLOCKED.** Stop, report `BLOCKED — AskUserQuestion unavailable`, and wait for the user. Do not write decisions to the plan file as a substitute, do not emit them as prose and stop, and do not silently auto-decide (only `/plan-tune` AUTO_DECIDE opt-ins authorize auto-picking).
 
 ### Format
 
-Every AskUserQuestion is a decision brief and must be sent as tool_use, not prose.
+Every AskUserQuestion decision has two parts: a markdown decision brief before the call, then a compact tool_use payload. Do not pack the full brief into the tool's `question` string. Every AskUserQuestion must be sent as tool_use, not prose.
 
 ```
 D<N> — <one-line question title>
@@ -324,6 +324,12 @@ Effort both-scales: when an option involves effort, label both human-team and CC
 
 Net line closes the tradeoff. Per-skill instructions may add stricter rules.
 
+Tool payload rules:
+- `question` is only the decision prompt: one sentence, no newlines, <=80 chars.
+- Background, regrounding, ELI10, stakes, recommendation, pros/cons, and trade-off tables stay in the markdown brief before the tool call.
+- Ask one decision per tool call when possible; batch at most two related questions/tabs. Sequence independent decisions instead of sending 3+ tabs.
+- Do not duplicate the same trade-off text in both `question` and `options[].description`. Prefer putting option-specific trade-offs in `options[].description`.
+
 12. **Non-ASCII characters — write directly, never \u-escape.** When any
     string field (question, option label, option description) contains
     Chinese (繁體/簡體), Japanese, Korean, or other non-ASCII text, emit
@@ -348,13 +354,17 @@ Net line closes the tradeoff. Per-skill instructions may add stricter rules.
 
 Before calling AskUserQuestion, verify:
 - [ ] D<N> header present
-- [ ] ELI10 paragraph present (stakes line too)
-- [ ] Recommendation line present with concrete reason
-- [ ] Completeness scored (coverage) OR kind-note present (kind)
+- [ ] ELI10 paragraph and stakes line present
+- [ ] Recommendation line present with reason
+- [ ] Completeness scored OR kind-note present
 - [ ] Every option has ≥2 ✅ and ≥1 ❌, each ≥40 chars (or hard-stop escape)
 - [ ] (recommended) label on one option (even for neutral-posture)
 - [ ] Dual-scale effort labels on effort-bearing options (human / CC)
 - [ ] Net line closes the decision
+- [ ] `question` is one sentence, no newlines, <=80 chars
+- [ ] Tool call has no more than two related questions/tabs
+- [ ] No duplicated trade-off text between `question` and `options[].description`
+- [ ] You wrote the brief, then called the tool_use payload
 - [ ] You are calling the tool, not writing prose
 - [ ] Non-ASCII characters (CJK / accents) written directly, NOT \u-escaped
 

canary/SKILL.md

@@ -275,15 +275,15 @@ AI orchestrator (e.g., OpenClaw). In spawned sessions:
 
 ### Tool resolution (read first)
 
-"AskUserQuestion" can resolve to two tools at runtime: the **host MCP variant** (e.g. `mcp__conductor__AskUserQuestion` — appears in your tool list when the host registers it) or the **native** Claude Code tool.
+"AskUserQuestion" may be host MCP (e.g. `mcp__conductor__AskUserQuestion`) or native.
 
-**Rule:** if any `mcp__*__AskUserQuestion` variant is in your tool list, prefer it. Hosts may disable native AUQ via `--disallowedTools AskUserQuestion` (Conductor does, by default) and route through their MCP variant; calling native there silently fails. Same questions/options shape; same decision-brief format applies.
+**Rule:** if any `mcp__*__AskUserQuestion` variant is in your tool list, prefer it. Hosts may disable native AUQ and route through MCP; same shape and brief.
 
 **If no AskUserQuestion variant appears in your tool list, this skill is BLOCKED.** Stop, report `BLOCKED — AskUserQuestion unavailable`, and wait for the user. Do not write decisions to the plan file as a substitute, do not emit them as prose and stop, and do not silently auto-decide (only `/plan-tune` AUTO_DECIDE opt-ins authorize auto-picking).
 
 ### Format
 
-Every AskUserQuestion is a decision brief and must be sent as tool_use, not prose.
+Every AskUserQuestion decision has two parts: a markdown decision brief before the call, then a compact tool_use payload. Do not pack the full brief into the tool's `question` string. Every AskUserQuestion must be sent as tool_use, not prose.
 
 ```
 D<N> — <one-line question title>
@@ -316,6 +316,12 @@ Effort both-scales: when an option involves effort, label both human-team and CC
 
 Net line closes the tradeoff. Per-skill instructions may add stricter rules.
 
+Tool payload rules:
+- `question` is only the decision prompt: one sentence, no newlines, <=80 chars.
+- Background, regrounding, ELI10, stakes, recommendation, pros/cons, and trade-off tables stay in the markdown brief before the tool call.
+- Ask one decision per tool call when possible; batch at most two related questions/tabs. Sequence independent decisions instead of sending 3+ tabs.
+- Do not duplicate the same trade-off text in both `question` and `options[].description`. Prefer putting option-specific trade-offs in `options[].description`.
+
 12. **Non-ASCII characters — write directly, never \u-escape.** When any
     string field (question, option label, option description) contains
     Chinese (繁體/簡體), Japanese, Korean, or other non-ASCII text, emit
@@ -340,13 +346,17 @@ Net line closes the tradeoff. Per-skill instructions may add stricter rules.
 
 Before calling AskUserQuestion, verify:
 - [ ] D<N> header present
-- [ ] ELI10 paragraph present (stakes line too)
-- [ ] Recommendation line present with concrete reason
-- [ ] Completeness scored (coverage) OR kind-note present (kind)
+- [ ] ELI10 paragraph and stakes line present
+- [ ] Recommendation line present with reason
+- [ ] Completeness scored OR kind-note present
 - [ ] Every option has ≥2 ✅ and ≥1 ❌, each ≥40 chars (or hard-stop escape)
 - [ ] (recommended) label on one option (even for neutral-posture)
 - [ ] Dual-scale effort labels on effort-bearing options (human / CC)
 - [ ] Net line closes the decision
+- [ ] `question` is one sentence, no newlines, <=80 chars
+- [ ] Tool call has no more than two related questions/tabs
+- [ ] No duplicated trade-off text between `question` and `options[].description`
+- [ ] You wrote the brief, then called the tool_use payload
 - [ ] You are calling the tool, not writing prose
 - [ ] Non-ASCII characters (CJK / accents) written directly, NOT \u-escaped
 

codex/SKILL.md

@@ -277,15 +277,15 @@ AI orchestrator (e.g., OpenClaw). In spawned sessions:
 
 ### Tool resolution (read first)
 
-"AskUserQuestion" can resolve to two tools at runtime: the **host MCP variant** (e.g. `mcp__conductor__AskUserQuestion` — appears in your tool list when the host registers it) or the **native** Claude Code tool.
+"AskUserQuestion" may be host MCP (e.g. `mcp__conductor__AskUserQuestion`) or native.
 
-**Rule:** if any `mcp__*__AskUserQuestion` variant is in your tool list, prefer it. Hosts may disable native AUQ via `--disallowedTools AskUserQuestion` (Conductor does, by default) and route through their MCP variant; calling native there silently fails. Same questions/options shape; same decision-brief format applies.
+**Rule:** if any `mcp__*__AskUserQuestion` variant is in your tool list, prefer it. Hosts may disable native AUQ and route through MCP; same shape and brief.
 
 **If no AskUserQuestion variant appears in your tool list, this skill is BLOCKED.** Stop, report `BLOCKED — AskUserQuestion unavailable`, and wait for the user. Do not write decisions to the plan file as a substitute, do not emit them as prose and stop, and do not silently auto-decide (only `/plan-tune` AUTO_DECIDE opt-ins authorize auto-picking).
 
 ### Format
 
-Every AskUserQuestion is a decision brief and must be sent as tool_use, not prose.
+Every AskUserQuestion decision has two parts: a markdown decision brief before the call, then a compact tool_use payload. Do not pack the full brief into the tool's `question` string. Every AskUserQuestion must be sent as tool_use, not prose.
 
 ```
 D<N> — <one-line question title>
@@ -318,6 +318,12 @@ Effort both-scales: when an option involves effort, label both human-team and CC
 
 Net line closes the tradeoff. Per-skill instructions may add stricter rules.
 
+Tool payload rules:
+- `question` is only the decision prompt: one sentence, no newlines, <=80 chars.
+- Background, regrounding, ELI10, stakes, recommendation, pros/cons, and trade-off tables stay in the markdown brief before the tool call.
+- Ask one decision per tool call when possible; batch at most two related questions/tabs. Sequence independent decisions instead of sending 3+ tabs.
+- Do not duplicate the same trade-off text in both `question` and `options[].description`. Prefer putting option-specific trade-offs in `options[].description`.
+
 12. **Non-ASCII characters — write directly, never \u-escape.** When any
     string field (question, option label, option description) contains
     Chinese (繁體/簡體), Japanese, Korean, or other non-ASCII text, emit
@@ -342,13 +348,17 @@ Net line closes the tradeoff. Per-skill instructions may add stricter rules.
 
 Before calling AskUserQuestion, verify:
 - [ ] D<N> header present
-- [ ] ELI10 paragraph present (stakes line too)
-- [ ] Recommendation line present with concrete reason
-- [ ] Completeness scored (coverage) OR kind-note present (kind)
+- [ ] ELI10 paragraph and stakes line present
+- [ ] Recommendation line present with reason
+- [ ] Completeness scored OR kind-note present
 - [ ] Every option has ≥2 ✅ and ≥1 ❌, each ≥40 chars (or hard-stop escape)
 - [ ] (recommended) label on one option (even for neutral-posture)
 - [ ] Dual-scale effort labels on effort-bearing options (human / CC)
 - [ ] Net line closes the decision
+- [ ] `question` is one sentence, no newlines, <=80 chars
+- [ ] Tool call has no more than two related questions/tabs
+- [ ] No duplicated trade-off text between `question` and `options[].description`
+- [ ] You wrote the brief, then called the tool_use payload
 - [ ] You are calling the tool, not writing prose
 - [ ] Non-ASCII characters (CJK / accents) written directly, NOT \u-escaped
 

context-restore/SKILL.md

@@ -279,15 +279,15 @@ AI orchestrator (e.g., OpenClaw). In spawned sessions:
 
 ### Tool resolution (read first)
 
-"AskUserQuestion" can resolve to two tools at runtime: the **host MCP variant** (e.g. `mcp__conductor__AskUserQuestion` — appears in your tool list when the host registers it) or the **native** Claude Code tool.
+"AskUserQuestion" may be host MCP (e.g. `mcp__conductor__AskUserQuestion`) or native.
 
-**Rule:** if any `mcp__*__AskUserQuestion` variant is in your tool list, prefer it. Hosts may disable native AUQ via `--disallowedTools AskUserQuestion` (Conductor does, by default) and route through their MCP variant; calling native there silently fails. Same questions/options shape; same decision-brief format applies.
+**Rule:** if any `mcp__*__AskUserQuestion` variant is in your tool list, prefer it. Hosts may disable native AUQ and route through MCP; same shape and brief.
 
 **If no AskUserQuestion variant appears in your tool list, this skill is BLOCKED.** Stop, report `BLOCKED — AskUserQuestion unavailable`, and wait for the user. Do not write decisions to the plan file as a substitute, do not emit them as prose and stop, and do not silently auto-decide (only `/plan-tune` AUTO_DECIDE opt-ins authorize auto-picking).
 
 ### Format
 
-Every AskUserQuestion is a decision brief and must be sent as tool_use, not prose.
+Every AskUserQuestion decision has two parts: a markdown decision brief before the call, then a compact tool_use payload. Do not pack the full brief into the tool's `question` string. Every AskUserQuestion must be sent as tool_use, not prose.
 
 ```
 D<N> — <one-line question title>
@@ -320,6 +320,12 @@ Effort both-scales: when an option involves effort, label both human-team and CC
 
 Net line closes the tradeoff. Per-skill instructions may add stricter rules.
 
+Tool payload rules:
+- `question` is only the decision prompt: one sentence, no newlines, <=80 chars.
+- Background, regrounding, ELI10, stakes, recommendation, pros/cons, and trade-off tables stay in the markdown brief before the tool call.
+- Ask one decision per tool call when possible; batch at most two related questions/tabs. Sequence independent decisions instead of sending 3+ tabs.
+- Do not duplicate the same trade-off text in both `question` and `options[].description`. Prefer putting option-specific trade-offs in `options[].description`.
+
 12. **Non-ASCII characters — write directly, never \u-escape.** When any
     string field (question, option label, option description) contains
     Chinese (繁體/簡體), Japanese, Korean, or other non-ASCII text, emit
@@ -344,13 +350,17 @@ Net line closes the tradeoff. Per-skill instructions may add stricter rules.
 
 Before calling AskUserQuestion, verify:
 - [ ] D<N> header present
-- [ ] ELI10 paragraph present (stakes line too)
-- [ ] Recommendation line present with concrete reason
-- [ ] Completeness scored (coverage) OR kind-note present (kind)
+- [ ] ELI10 paragraph and stakes line present
+- [ ] Recommendation line present with reason
+- [ ] Completeness scored OR kind-note present
 - [ ] Every option has ≥2 ✅ and ≥1 ❌, each ≥40 chars (or hard-stop escape)
 - [ ] (recommended) label on one option (even for neutral-posture)
 - [ ] Dual-scale effort labels on effort-bearing options (human / CC)
 - [ ] Net line closes the decision
+- [ ] `question` is one sentence, no newlines, <=80 chars
+- [ ] Tool call has no more than two related questions/tabs
+- [ ] No duplicated trade-off text between `question` and `options[].description`
+- [ ] You wrote the brief, then called the tool_use payload
 - [ ] You are calling the tool, not writing prose
 - [ ] Non-ASCII characters (CJK / accents) written directly, NOT \u-escaped
 

context-save/SKILL.md

@@ -279,15 +279,15 @@ AI orchestrator (e.g., OpenClaw). In spawned sessions:
 
 ### Tool resolution (read first)
 
-"AskUserQuestion" can resolve to two tools at runtime: the **host MCP variant** (e.g. `mcp__conductor__AskUserQuestion` — appears in your tool list when the host registers it) or the **native** Claude Code tool.
+"AskUserQuestion" may be host MCP (e.g. `mcp__conductor__AskUserQuestion`) or native.
 
-**Rule:** if any `mcp__*__AskUserQuestion` variant is in your tool list, prefer it. Hosts may disable native AUQ via `--disallowedTools AskUserQuestion` (Conductor does, by default) and route through their MCP variant; calling native there silently fails. Same questions/options shape; same decision-brief format applies.
+**Rule:** if any `mcp__*__AskUserQuestion` variant is in your tool list, prefer it. Hosts may disable native AUQ and route through MCP; same shape and brief.
 
 **If no AskUserQuestion variant appears in your tool list, this skill is BLOCKED.** Stop, report `BLOCKED — AskUserQuestion unavailable`, and wait for the user. Do not write decisions to the plan file as a substitute, do not emit them as prose and stop, and do not silently auto-decide (only `/plan-tune` AUTO_DECIDE opt-ins authorize auto-picking).
 
 ### Format
 
-Every AskUserQuestion is a decision brief and must be sent as tool_use, not prose.
+Every AskUserQuestion decision has two parts: a markdown decision brief before the call, then a compact tool_use payload. Do not pack the full brief into the tool's `question` string. Every AskUserQuestion must be sent as tool_use, not prose.
 
 ```
 D<N> — <one-line question title>
@@ -320,6 +320,12 @@ Effort both-scales: when an option involves effort, label both human-team and CC
 
 Net line closes the tradeoff. Per-skill instructions may add stricter rules.
 
+Tool payload rules:
+- `question` is only the decision prompt: one sentence, no newlines, <=80 chars.
+- Background, regrounding, ELI10, stakes, recommendation, pros/cons, and trade-off tables stay in the markdown brief before the tool call.
+- Ask one decision per tool call when possible; batch at most two related questions/tabs. Sequence independent decisions instead of sending 3+ tabs.
+- Do not duplicate the same trade-off text in both `question` and `options[].description`. Prefer putting option-specific trade-offs in `options[].description`.
+
 12. **Non-ASCII characters — write directly, never \u-escape.** When any
     string field (question, option label, option description) contains
     Chinese (繁體/簡體), Japanese, Korean, or other non-ASCII text, emit
@@ -344,13 +350,17 @@ Net line closes the tradeoff. Per-skill instructions may add stricter rules.
 
 Before calling AskUserQuestion, verify:
 - [ ] D<N> header present
-- [ ] ELI10 paragraph present (stakes line too)
-- [ ] Recommendation line present with concrete reason
-- [ ] Completeness scored (coverage) OR kind-note present (kind)
+- [ ] ELI10 paragraph and stakes line present
+- [ ] Recommendation line present with reason
+- [ ] Completeness scored OR kind-note present
 - [ ] Every option has ≥2 ✅ and ≥1 ❌, each ≥40 chars (or hard-stop escape)
 - [ ] (recommended) label on one option (even for neutral-posture)
 - [ ] Dual-scale effort labels on effort-bearing options (human / CC)
 - [ ] Net line closes the decision
+- [ ] `question` is one sentence, no newlines, <=80 chars
+- [ ] Tool call has no more than two related questions/tabs
+- [ ] No duplicated trade-off text between `question` and `options[].description`
+- [ ] You wrote the brief, then called the tool_use payload
 - [ ] You are calling the tool, not writing prose
 - [ ] Non-ASCII characters (CJK / accents) written directly, NOT \u-escaped