diff --git a/dashboard.html b/dashboard.html
index 899ecbb..5cdfee6 100644
--- a/dashboard.html
+++ b/dashboard.html
@@ -283,7 +283,21 @@
 #copy-checkin{background:var(--accent);color:var(--bg);border:0;border-radius:6px;padding:10px 16px;cursor:pointer;font-weight:600}
 #copy-status{margin-left:12px}
 code{color:var(--accent);word-break:break-all}
-pre{white-space:pre-wrap;overflow-wrap:anywhere}details{margin:1rem 0}dd{margin-left:1rem}</style></head><body><header class="hero"><span class="orb"><svg class="mark" viewBox="0 0 48 48" fill="none" stroke="currentColor" stroke-width="1.3" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><circle cx="24" cy="20" r="13"/><path d="M18,32 L18,38 L30,38 L30,32 M20,43 L28,43 M24,7 L24,20 L31,24"/></svg></span><h1>PyAuto<b>Insight</b><span class="kind">Inference dashboard</span></h1><div class="rule"></div><p class="tag">Infer. Inspect. Understand.</p></header><nav class="board-nav" aria-label="Board sections"><a class="board-nav-card" href="#campaigns"><span class="board-nav-label">Active campaigns</span></a><a class="board-nav-card" href="#evidence"><span class="board-nav-label">Inference evidence</span></a><a class="board-nav-card" href="#lens"><span class="board-nav-label">Lens project</span></a></nav><main><!-- insight-inputs:d6d8f9578ff76984115694de64eb795549983d279fa228a2aa01f568bac14267 --><!-- insight-campaigns:cebf9dfa86c60986e2ae3381ae314259f7e491233af2ec497ff330f710f3e901 --><section class="orchestration-panel" id="orchestration-insight" aria-labelledby="orchestration-insight-heading" data-orchestration-panel><div class="orchestration-head"><div><h2 class="prompt-heading" id="orchestration-insight-heading">Find your next <strong>Insight</strong></h2></div><nav class="orchestration-links" aria-label="Work on GitHub"><a href="https://github.com/PyAutoLabs/PyAutoInsight">PyAutoInsight</a><a href="https://github.com/PyAutoLabs/autolens_inference">autolens_inference</a></nav></div><div class="orchestration-controls"><div class="orchestration-direction"><label for="orchestration-insight-direction">Optional direction</label><textarea id="orchestration-insight-direction" rows="2" data-orchestration-direction placeholder="Focus on a task, project or question"></textarea></div><button type="button" class="orchestration-copy" data-orchestration-copy>Copy check-in prompt</button></div><details data-orchestration-preview><summary>Read the prompt</summary><label class="sr-only" for="orchestration-insight-prompt">Exact prompt to copy</label><textarea id="orchestration-insight-prompt" data-orchestration-prompt readonly rows="8">Use PyAutoInsight as the home for all inference work in this chat. Read PyAutoInsight/AGENTS.md and CHECKIN.md, then campaigns.yaml, the linked tasks and current project campaign ledgers. Check every open campaign and task for updates since the last check-in: results, jobs, PRs, releases, blockers and the next bounded step. Distinguish verified updates from stale or unavailable evidence; do not infer success from missing data. Give me one concise campaign-by-campaign summary and a proposed priority order. Apply any campaign direction or ideas I supply before or after this prompt; otherwise cover everything. Keep the overall check-in even when I focus one campaign. Update the Insight ledger with verified facts and dated source links, regenerate the board, and use the repository workflow to persist changes. Use Cortex for scientific observations and human conclusions, and the existing Brain samplers faculty for optional advice and project drivers for execution. Do not launch new compute, change defaults or baselines, or bypass a human/release gate merely to check in. Continue managing subsequent inference requests in this chat. Optional direction: [leave blank, name a campaign, or suggest an idea].
+pre{white-space:pre-wrap;overflow-wrap:anywhere}details{margin:1rem 0}dd{margin-left:1rem}</style></head><body><header class="hero"><span class="orb"><svg class="mark" viewBox="0 0 48 48" fill="none" stroke="currentColor" stroke-width="1.3" stroke-linecap="round" stroke-linejoin="round" aria-hidden="true"><circle cx="24" cy="20" r="13"/><path d="M18,32 L18,38 L30,38 L30,32 M20,43 L28,43 M24,7 L24,20 L31,24"/></svg></span><h1>PyAuto<b>Insight</b><span class="kind">Inference dashboard</span></h1><div class="rule"></div><p class="tag">Infer. Inspect. Understand.</p></header><nav class="board-nav" aria-label="Board sections"><a class="board-nav-card" href="#campaigns"><span class="board-nav-label">Active campaigns</span></a><a class="board-nav-card" href="#evidence"><span class="board-nav-label">Inference evidence</span></a><a class="board-nav-card" href="#lens"><span class="board-nav-label">Lens project</span></a></nav><main><!-- insight-inputs:d6d8f9578ff76984115694de64eb795549983d279fa228a2aa01f568bac14267 --><!-- insight-campaigns:cebf9dfa86c60986e2ae3381ae314259f7e491233af2ec497ff330f710f3e901 --><section class="orchestration-panel" id="orchestration-insight" aria-labelledby="orchestration-insight-heading" data-orchestration-panel><div class="orchestration-head"><div><h2 class="prompt-heading" id="orchestration-insight-heading">Find your next <strong>Insight</strong></h2></div><nav class="orchestration-links" aria-label="Work on GitHub"><a href="https://github.com/PyAutoLabs/PyAutoInsight">PyAutoInsight</a><a href="https://github.com/PyAutoLabs/autolens_inference">autolens_inference</a></nav></div><div class="orchestration-controls"><div class="orchestration-direction"><label for="orchestration-insight-direction">Optional direction</label><textarea id="orchestration-insight-direction" rows="2" data-orchestration-direction placeholder="Focus on a task, project or question"></textarea></div><button type="button" class="orchestration-copy" data-orchestration-copy>Copy check-in prompt</button></div><details data-orchestration-preview><summary>Read the prompt</summary><label class="sr-only" for="orchestration-insight-prompt">Exact prompt to copy</label><textarea id="orchestration-insight-prompt" data-orchestration-prompt readonly rows="8">Use PyAutoInsight as the home for inference work in this ongoing chat. Read PyAutoInsight/AGENTS.md and CHECKIN.md, then the campaign ledger, relevant tasks and registered project evidence. Use the existing Brain samplers faculty when advice on sampler coverage or selection would help, and project-owned drivers for execution.
+
+When I give no particular direction, review every open campaign and task for changes since the last check-in: results, jobs, PRs, releases, blockers and recorded next steps. Give me a concise campaign-by-campaign summary and a proposed priority order. Distinguish verified updates from stale, missing or unavailable evidence.
+
+When I name a campaign, sampler, result or idea, make that the main focus. Help me inspect diagnostics, understand an unsuccessful run, compare inference methods, identify missing evidence, design a benchmark or develop a new campaign. Bring in related work where useful; do not repeat the full campaign review on every follow-up.
+
+Assess comparisons against the campaign’s stated objectives and acceptance criteria. Consider convergence, sampling quality, agreement with reference results, robustness and computational cost where relevant. Make differences in models, priors, datasets, hardware, precision, stopping rules and resource budgets explicit. Do not rank incompatible runs or treat speed alone as success.
+
+Discuss proposed experiments with me, explaining what they would establish and the resources they need. Help turn agreed direction into concrete campaign tasks. Keep inference intent and pending domain work in Insight, execution and raw results in project repositories, scientific observations and my conclusions in Cortex, and bounded implementation work in Mind.
+
+Update the Insight ledger with verified facts and dated source links, regenerate the board and persist changes through the repository workflow. Keep review dates separate from evidence freshness. Distinguish completed execution from scientific acceptance, and proposed interpretations from conclusions I have accepted.
+
+Carry clearly authorized work through the appropriate procedure, retaining decisions and approvals already given in this conversation. Launch compute or change defaults, baselines or campaign direction only when authorized; a general check-in does not authorize those actions.
+
+After taking action, report what changed, what the evidence supports and what remains unresolved. Continue subsequent inference work in this chat and stop at the session deliverable without scheduling background follow-up.
 
 Work on GitHub:
 - PyAutoInsight: https://github.com/PyAutoLabs/PyAutoInsight
diff --git a/dashboard.md b/dashboard.md
index 58ba375..668570b 100644
--- a/dashboard.md
+++ b/dashboard.md
@@ -16,7 +16,21 @@
 Copy this into one chat. Add a campaign focus or idea before or after it, or leave it unchanged.
 
 ```text
-Use PyAutoInsight as the home for all inference work in this chat. Read PyAutoInsight/AGENTS.md and CHECKIN.md, then campaigns.yaml, the linked tasks and current project campaign ledgers. Check every open campaign and task for updates since the last check-in: results, jobs, PRs, releases, blockers and the next bounded step. Distinguish verified updates from stale or unavailable evidence; do not infer success from missing data. Give me one concise campaign-by-campaign summary and a proposed priority order. Apply any campaign direction or ideas I supply before or after this prompt; otherwise cover everything. Keep the overall check-in even when I focus one campaign. Update the Insight ledger with verified facts and dated source links, regenerate the board, and use the repository workflow to persist changes. Use Cortex for scientific observations and human conclusions, and the existing Brain samplers faculty for optional advice and project drivers for execution. Do not launch new compute, change defaults or baselines, or bypass a human/release gate merely to check in. Continue managing subsequent inference requests in this chat. Optional direction: [leave blank, name a campaign, or suggest an idea].
+Use PyAutoInsight as the home for inference work in this ongoing chat. Read PyAutoInsight/AGENTS.md and CHECKIN.md, then the campaign ledger, relevant tasks and registered project evidence. Use the existing Brain samplers faculty when advice on sampler coverage or selection would help, and project-owned drivers for execution.
+
+When I give no particular direction, review every open campaign and task for changes since the last check-in: results, jobs, PRs, releases, blockers and recorded next steps. Give me a concise campaign-by-campaign summary and a proposed priority order. Distinguish verified updates from stale, missing or unavailable evidence.
+
+When I name a campaign, sampler, result or idea, make that the main focus. Help me inspect diagnostics, understand an unsuccessful run, compare inference methods, identify missing evidence, design a benchmark or develop a new campaign. Bring in related work where useful; do not repeat the full campaign review on every follow-up.
+
+Assess comparisons against the campaign’s stated objectives and acceptance criteria. Consider convergence, sampling quality, agreement with reference results, robustness and computational cost where relevant. Make differences in models, priors, datasets, hardware, precision, stopping rules and resource budgets explicit. Do not rank incompatible runs or treat speed alone as success.
+
+Discuss proposed experiments with me, explaining what they would establish and the resources they need. Help turn agreed direction into concrete campaign tasks. Keep inference intent and pending domain work in Insight, execution and raw results in project repositories, scientific observations and my conclusions in Cortex, and bounded implementation work in Mind.
+
+Update the Insight ledger with verified facts and dated source links, regenerate the board and persist changes through the repository workflow. Keep review dates separate from evidence freshness. Distinguish completed execution from scientific acceptance, and proposed interpretations from conclusions I have accepted.
+
+Carry clearly authorized work through the appropriate procedure, retaining decisions and approvals already given in this conversation. Launch compute or change defaults, baselines or campaign direction only when authorized; a general check-in does not authorize those actions.
+
+After taking action, report what changed, what the evidence supports and what remains unresolved. Continue subsequent inference work in this chat and stop at the session deliverable without scheduling background follow-up.
 ```
 
 Last check-in: not recorded yet. Ledger dates are review dates, not measurement freshness.
diff --git a/insight/campaigns.py b/insight/campaigns.py
index c7cea68..9da5ac6 100644
--- a/insight/campaigns.py
+++ b/insight/campaigns.py
@@ -28,7 +28,41 @@
     "superseded",
 }
 CLOSED = {"complete", "superseded"}
-PROMPT = """Use PyAutoInsight as the home for all inference work in this chat. Read PyAutoInsight/AGENTS.md and CHECKIN.md, then campaigns.yaml, the linked tasks and current project campaign ledgers. Check every open campaign and task for updates since the last check-in: results, jobs, PRs, releases, blockers and the next bounded step. Distinguish verified updates from stale or unavailable evidence; do not infer success from missing data. Give me one concise campaign-by-campaign summary and a proposed priority order. Apply any campaign direction or ideas I supply before or after this prompt; otherwise cover everything. Keep the overall check-in even when I focus one campaign. Update the Insight ledger with verified facts and dated source links, regenerate the board, and use the repository workflow to persist changes. Use Cortex for scientific observations and human conclusions, and the existing Brain samplers faculty for optional advice and project drivers for execution. Do not launch new compute, change defaults or baselines, or bypass a human/release gate merely to check in. Continue managing subsequent inference requests in this chat. Optional direction: [leave blank, name a campaign, or suggest an idea]."""
+PROMPT = (
+    "Use PyAutoInsight as the home for inference work in this ongoing chat. Read "
+    "PyAutoInsight/AGENTS.md and CHECKIN.md, then the campaign ledger, relevant tasks and "
+    "registered project evidence. Use the existing Brain samplers faculty when advice on "
+    "sampler coverage or selection would help, and project-owned drivers for execution.\n\n"
+    "When I give no particular direction, review every open campaign and task for changes "
+    "since the last check-in: results, jobs, PRs, releases, blockers and recorded next steps. "
+    "Give me a concise campaign-by-campaign summary and a proposed priority order. "
+    "Distinguish verified updates from stale, missing or unavailable evidence.\n\n"
+    "When I name a campaign, sampler, result or idea, make that the main focus. Help me "
+    "inspect diagnostics, understand an unsuccessful run, compare inference methods, identify "
+    "missing evidence, design a benchmark or develop a new campaign. Bring in related work "
+    "where useful; do not repeat the full campaign review on every follow-up.\n\n"
+    "Assess comparisons against the campaign’s stated objectives and acceptance criteria. "
+    "Consider convergence, sampling quality, agreement with reference results, robustness and "
+    "computational cost where relevant. Make differences in models, priors, datasets, "
+    "hardware, precision, stopping rules and resource budgets explicit. Do not rank "
+    "incompatible runs or treat speed alone as success.\n\n"
+    "Discuss proposed experiments with me, explaining what they would establish and the "
+    "resources they need. Help turn agreed direction into concrete campaign tasks. Keep "
+    "inference intent and pending domain work in Insight, execution and raw results in "
+    "project repositories, scientific observations and my conclusions in Cortex, and bounded "
+    "implementation work in Mind.\n\n"
+    "Update the Insight ledger with verified facts and dated source links, regenerate the "
+    "board and persist changes through the repository workflow. Keep review dates separate "
+    "from evidence freshness. Distinguish completed execution from scientific acceptance, and "
+    "proposed interpretations from conclusions I have accepted.\n\n"
+    "Carry clearly authorized work through the appropriate procedure, retaining decisions and "
+    "approvals already given in this conversation. Launch compute or change defaults, "
+    "baselines or campaign direction only when authorized; a general check-in does not "
+    "authorize those actions.\n\n"
+    "After taking action, report what changed, what the evidence supports and what remains "
+    "unresolved. Continue subsequent inference work in this chat and stop at the session "
+    "deliverable without scheduling background follow-up."
+)
 
 
 class CampaignError(ValueError):