@piagent/platform 1.2.1 → 1.2.3

package/packages/piagent-core/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@piagent/core",
-  "version": "1.2.1",
+  "version": "1.2.3",
   "type": "module",
   "private": true,
   "license": "MIT",

package/packages/piagent-core/prompts/be-to-fe.md

@@ -20,11 +20,11 @@ Use this for tasks like:
 
 Mandatory flow:
 
-0. Call `piagent_context_preflight` with `workflow=be-to-fe`. If it recommends `fresh-session`, stop loading context in this session and tell the user to use `/fresh-be-to-fe <request>` unless this command already runs in a fresh session.
+0. Call `piagent_context_preflight` with `workflow=be-to-fe`. If it recommends `fresh-session`, stop loading context in this session and tell the user to use `/piagent-session fresh be-to-fe <request>` unless this command already runs in a fresh session.
 1. Call `piagent_context` with `detail=full` and confirm profile/protected backend paths.
 2. Call `piagent_orchestration_policy`; keep the workflow solo-first unless bounded BE/FE scouting or review is worth the extra token/tool cost.
 3. Call `piagent_memory_status`; search memory or Field Guide for prior BE/FE mapping decisions if relevant, record citations with `piagent_memory_citation_record`, then verify against current BE/FE files.
-4. Read `.pi/project-context.md`. If pending, stop and ask for `/onboard-project`.
+4. Read `.pi/project-context.md`. If pending, stop and ask for `/onboard-project run`.
 5. Read required context from the active profile and call `piagent_context_budget` for BE/FE files that look large.
 6. Classify the task:
    - BE scout: read-only.
@@ -63,10 +63,11 @@ Mandatory flow:
 16. Run the exact FE verify command from the profile/task through Pi bash and record the observed result with `piagent_verify_record`. Ad-hoc commands are advisory only and will not satisfy the passing gate.
 17. Record handoff with `piagent_trace_record`.
 18. Call `piagent_task_gate_check`; if it fails, report blocked/partial instead of done.
+19. Do not paste full test/build/tool logs into chat or final output. Summarize the signal, quote only relevant failing lines, and use `/piagent-logs` or a narrower command when oversized output has been compacted.
 
 For generic projects, use profile `be-readonly-fe` when the repo policy is “BE scout only, FE write allowed”.
 
-Do not ask the user to paste the mandatory flow. The platform prompt already contains it. If the user asks only for scout/audit, stay read-only and prefer `/scout` or `/fresh-scout`.
+Do not ask the user to paste the mandatory flow. The platform prompt already contains it. If the user asks only for scout/audit, stay read-only and prefer `/workflow scout` or `/piagent-session fresh scout`.
 
 Final output:
 

package/packages/piagent-core/prompts/discuss.md

@@ -13,7 +13,7 @@ Operating rules:
 
 1. Call `piagent_context` first when available.
 2. Call `piagent_memory_status`; use memory only to avoid repeated context questions, not as authority.
-3. Read `.pi/project-context.md` if available; if it is still pending, recommend `/onboard-project` before implementation.
+3. Read `.pi/project-context.md` if available; if it is still pending, recommend `/onboard-project run` before implementation.
 4. Inspect relevant files/docs before asking; do not ask what can be answered from the project.
 5. Identify the next unresolved decision, dependency, constraint, or risk.
 6. Ask at most 3 focused questions per round.
@@ -32,4 +32,4 @@ Final output when clear:
 - implementation approach
 - required context
 - verify commands
-- recommended next command: `/plan` or `/task`
+- recommended next command: `/workflow plan` or `/workflow task`

package/packages/piagent-core/prompts/plan.md

@@ -14,7 +14,7 @@ Rules:
 1. Call `piagent_context`.
 2. Call `piagent_orchestration_policy`; default to solo-first and choose review lenses/model roles before considering subagents.
 3. Call `piagent_memory_status`; search memory or Field Guide if it can reduce re-scouting, but treat it as advisory.
-4. Read `.pi/project-context.md`; if it is missing or still pending, recommend `/onboard-project` before implementation.
+4. Read `.pi/project-context.md`; if it is missing or still pending, recommend `/onboard-project run` before implementation.
 5. Read required context.
 6. Identify protected paths and external/high-risk actions.
 7. Decide whether subagents are useful enough to justify extra token/tool cost. If the bundled `pi-subagents` parent skill is available, use it for delegation patterns and safety boundaries. If `pi-subagents`/`subagent(...)` is available, use `piagent-scout` for independent read-only repo/spec mapping before medium/large plans. Use builtin `context-builder` when the task needs a handoff context/meta-prompt. Use `piagent-oracle` for architecture/risk challenge when the decision is non-obvious. Continue single-agent for tiny plans or unavailable subagent tooling.

package/packages/piagent-core/prompts/platform-improve.md

@@ -23,7 +23,7 @@ Mandatory flow:
 1. Call `piagent_context` with `detail=full` and inspect the active project profile.
 2. Call `piagent_orchestration_policy`; keep the platform change solo-first unless bounded scout/planning/review lanes are useful.
 3. Call `piagent_memory_status`; if prior decisions or Field Guide notes are relevant, search memory and record citations with `piagent_memory_citation_record`.
-4. Read `.pi/project-context.md`. If pending, stop and ask the user to run `/onboard-project`.
+4. Read `.pi/project-context.md`. If pending, stop and ask the user to run `/onboard-project run`.
 5. Read the required context declared by the active profile. Before loading large files, call `piagent_context_budget`.
 6. Classify the task risk lane and create a task contract with `piagent_task_start`, including review lenses and a compact workPlan, before source edits.
 7. Decide whether bounded subagents are useful enough to justify extra token/tool cost:
@@ -42,6 +42,7 @@ Mandatory flow:
 12. Implement only the bounded target behavior.
 13. If runtime behavior changes, update README/docs and add/adjust a decision note when appropriate.
 14. For source-changing tasks, run the exact verify command from `task.verifyCommands` through Pi bash, record observed verify evidence, trace, then call `piagent_task_gate_check` before final.
+15. Do not paste full test/build/tool logs into chat or final output. Summarize the signal, quote only relevant failing lines, and use `/piagent-logs` or a narrower command when oversized output has been compacted.
 
 Default verification:
 

package/packages/piagent-core/prompts/pr.md

@@ -16,7 +16,7 @@ Rules:
 1. Do not merge. Do not publish, release, deploy, or tag.
 2. Use normal shell Git/GitHub commands so the piagent guard and confirmation gates remain active.
 3. Start with `git status --short`, `git branch --show-current`, and `git remote -v`.
-4. If there are uncommitted changes, stop and ask the operator to run `/commit` first, unless the operator explicitly asks this command to include a commit.
+4. If there are uncommitted changes, stop and ask the operator to run `/workflow commit` first, unless the operator explicitly asks this command to include a commit.
 5. Run the relevant verification command before opening the PR. At minimum use `git diff --check`; for source or policy changes, run the project verify/test/typecheck commands that apply.
 6. If the branch is not pushed, ask for explicit operator confirmation before `git push -u origin <branch>`.
 7. Create the PR only after the operator explicitly confirms the external GitHub action. Prefer `gh pr create --draft` with a clear title/body derived from the current diff and commits, unless the operator explicitly asks for a ready-for-review PR.

package/packages/piagent-core/prompts/review.md

@@ -11,7 +11,7 @@ $ARGUMENTS
 
 Rules:
 
-0. Call `piagent_context_preflight` with `workflow=review` for large diffs or long requests. If it recommends `fresh-session`, prefer `/fresh-scout` for read-only evidence gathering or ask the user to resume in a fresh session.
+0. Call `piagent_context_preflight` with `workflow=review` for large diffs or long requests. If it recommends `fresh-session`, prefer `/piagent-session fresh scout` for read-only evidence gathering or ask the user to resume in a fresh session.
 1. Call `piagent_context`.
 2. Call `piagent_orchestration_policy` and select explicit review lenses before reading the diff deeply.
 3. Read `.pi/project-context.md` if available.
@@ -26,6 +26,7 @@ Rules:
    - security/high-risk only when relevant.
    If the user asks for a loop, use `/review-loop` semantics or equivalent parent-controlled max-round loop. Continue single-agent for tiny diffs or unavailable subagent tooling.
 9. Report findings by severity.
+10. Do not paste full diff/test/tool logs into chat or final output. Summarize the signal, quote only relevant lines, and use `/piagent-logs` or narrower commands when oversized output has been compacted.
 
 Output:
 

package/packages/piagent-core/prompts/scout.md

@@ -14,11 +14,11 @@ Use this when the user needs evidence before deciding whether to implement. Do n
 Preflight first:
 
 1. Call `piagent_context_preflight` with `workflow=scout`.
-2. If it recommends `fresh-session`, stop loading context in this session and tell the user to use `/fresh-scout <request>` unless this command already runs in a fresh session.
+2. If it recommends `fresh-session`, stop loading context in this session and tell the user to use `/piagent-session fresh scout <request>` unless this command already runs in a fresh session.
 3. Call `piagent_context` with `detail=full`.
 4. Call `piagent_orchestration_policy`; keep scout solo-first unless independent read-only lanes are useful.
 5. Call `piagent_memory_status`; cite relevant memory or Field Guide only with `piagent_memory_citation_record`, then verify against current repo files.
-6. Read `.pi/project-context.md`; if pending, stop and ask for `/onboard-project`.
+6. Read `.pi/project-context.md`; if pending, stop and ask for `/onboard-project run`.
 7. Create a read-only task contract with `piagent_task_start` before broad scouting:
    - risk lane;
    - expected output;
@@ -34,6 +34,7 @@ Preflight first:
 12. Run the exact verify command from the task contract and record with `piagent_verify_record`.
 13. Record handoff with `piagent_trace_record`.
 14. Call `piagent_task_gate_check`. If the gate fails, report partial/blocked.
+15. Do not paste full tool/test logs into chat or final output. Summarize the signal, quote only relevant lines, and use `/piagent-logs` or a narrower command when oversized output has been compacted.
 
 Output:
 

package/packages/piagent-core/prompts/task.md

@@ -11,11 +11,11 @@ $ARGUMENTS
 
 Mandatory flow:
 
-0. Call `piagent_context_preflight` with `workflow=task`. If it recommends `fresh-session`, stop loading context in this session and tell the user to use `/fresh-task <request>` unless this command already runs in a fresh session.
+0. Call `piagent_context_preflight` with `workflow=task`. If it recommends `fresh-session`, stop loading context in this session and tell the user to use `/piagent-session fresh task <request>` unless this command already runs in a fresh session.
 1. Call `piagent_context` and read the project profile/runtime policy.
 2. Call `piagent_orchestration_policy` and keep the task solo-first unless the policy and task shape make bounded subagents clearly useful.
 3. Call `piagent_memory_status`. If memory or the Field Guide is enabled and relevant to the task, search/read it as advisory context, record citations with `piagent_memory_citation_record`, then verify against current repo files.
-4. Call `piagent_context_index_status` when available, then read `.pi/project-context.md`. If `.pi/project-context.md` is missing or still says `Generated: not yet`, stop and ask the user to run `/onboard-project` after login/model selection before implementation.
+4. Call `piagent_context_index_status` when available, then read `.pi/project-context.md`. If `.pi/project-context.md` is missing or still says `Generated: not yet`, stop and ask the user to run `/onboard-project run` after login/model selection before implementation.
    - Use `piagent_context_index_search` for navigation hints when the task touches an unfamiliar module/tech/risk area.
    - Treat context-index hits as advisory; open and verify cited files before editing.
 5. Build a Task Implementation Contract with `piagent_task_start` before editing:
@@ -52,6 +52,7 @@ Mandatory flow:
 16. Call `piagent_task_gate_check`. If gate fails, final outcome is blocked/partial, not done.
 17. If the user asks about token/context/cost usage, call `piagent_usage_snapshot`; for exact token/cost totals, tell the user to run `/session` or `piagent-usage <project-path>`.
 18. If verify cannot run, stop and report the exact blocker. Do not call it done.
+19. Do not paste full test/build/tool logs into chat or final output. Summarize the signal, quote only the relevant failing lines, and use `/piagent-logs` or a narrower command when oversized output has been compacted.
 
 Do not ask the user to paste this mandatory flow. The platform prompt already contains it. If the user pasted the full flow, treat it as boilerplate and extract only the task request.
 

package/scripts/init-project.sh

@@ -315,10 +315,11 @@ echo "Next:"
 echo "  cd \"$PROJECT_PATH\""
 echo "  pi"
 echo "  /login              # first time only"
-echo "  /onboard-project    # first project-read after model selection"
+echo "  /onboard    # open onboarding menu/status"
+echo "  /onboard run # first project-read after model selection"
 echo "  /mcp                # MCP servers come from the global install, not from this project"
-echo "  /context-index      # inspect compact project context index"
-echo "  /memory-policy      # inspect project memory policy when needed"
+echo "  /context index # inspect compact project context index"
+echo "  /memory     # inspect project memory policy when needed"
 echo
 echo "Note: .mcp.json and .pi/mcp.json are written empty on purpose. They exist so this"
 echo "project can add its own servers later; the shared baseline lives in the global"

package/scripts/install-global.sh

@@ -178,6 +178,50 @@ run_cmd() {
   fi
 }
 
+resolve_user_package_path() {
+  local source="$1"
+  local pi_agent_dir="${PI_CODING_AGENT_DIR:-"${HOME}/.pi/agent"}"
+  node --input-type=module - "$pi_agent_dir" "$source" <<'NODE'
+import path from "node:path";
+
+const base = process.argv[2];
+const source = process.argv[3];
+console.log(path.resolve(base, source));
+NODE
+}
+
+is_piagent_platform_source() {
+  local source="$1"
+  case "$source" in
+    git:github.com/Vt-mmm/piagent|git:github.com/Vt-mmm/piagent@*|https://github.com/Vt-mmm/piagent*|npm:@piagent/platform|npm:@piagent/platform@*)
+      return 0
+      ;;
+    npm:*|git:*|http:*|https:*)
+      return 1
+      ;;
+  esac
+
+  local resolved_source
+  resolved_source="$(resolve_user_package_path "$source" 2>/dev/null || true)"
+  [[ "$resolved_source" == "$PLATFORM_ROOT" ]]
+}
+
+remove_existing_piagent_platform_sources() {
+  local source
+  while IFS= read -r source; do
+    [[ -n "$source" ]] || continue
+    if is_piagent_platform_source "$source"; then
+      run_cmd pi remove "$source"
+    fi
+  done < <(
+    pi list --approve 2>/dev/null | awk '
+      $0 == "User packages:" { in_user = 1; next }
+      $0 == "Project packages:" { in_user = 0; next }
+      in_user && $0 ~ /^  [^ ]/ { sub(/^  /, ""); print }
+    '
+  )
+}
+
 verify_npm_integrity() {
   local source="$1"
   local expected="$2"
@@ -496,6 +540,8 @@ if [[ -n "$RESOLVED_PACKAGE_COMMIT" ]]; then
   echo "  resolvedCommit: $RESOLVED_PACKAGE_COMMIT"
 fi
 echo "  source: $PACKAGE_SOURCE"
+echo "Removing previous Pi Agent Platform package registrations:"
+remove_existing_piagent_platform_sources
 run_cmd pi install "$PACKAGE_SOURCE"
 
 if [[ "$WITH_MCP" == true ]]; then
@@ -562,4 +608,4 @@ if [[ "$WITH_MCP" == true ]]; then
   echo "  /mcp            # inspect MCP servers; authenticate Figma/GitHub only when needed"
 fi
 echo "  /subagents-doctor"
-echo "  /task Implement <task>  # parent may auto-delegate scout/planner/reviewer"
+echo "  /workflow task Implement <task>  # parent may auto-delegate scout/planner/reviewer"

package/scripts/profile-doctor.sh

@@ -198,7 +198,7 @@ if (contextIndex.enabled && contextIndex.writePolicy !== "off") {
   if (relativeIndexPath.startsWith("..") || path.isAbsolute(relativeIndexPath)) {
     errors.push(`contextIndex.path escapes project root: ${contextIndex.path}`);
   } else if (!fs.existsSync(indexPath)) {
-    warnings.push(`context index missing: ${contextIndex.path}; run /onboard-project or record piagent_context_index_record`);
+    warnings.push(`context index missing: ${contextIndex.path}; run /onboard run or record piagent_context_index_record`);
   }
 }
 

package/scripts/setup.sh

@@ -452,12 +452,13 @@ if [[ "$DO_PROJECT" == true ]]; then
   echo "  /scoped-models     # optional: edit Ctrl+P model cycle scope"
   echo "  /mcp               # inspect MCP servers; run /mcp setup for guided changes"
   echo "  /subagents-doctor  # inspect subagent setup"
-  echo "  /onboard-project   # first project-read snapshot before implementation"
-  echo "  /memory-policy     # inspect project memory policy when needed"
+  echo "  /onboard   # open onboarding menu/status"
+  echo "  /onboard run # first project-read snapshot before implementation"
+  echo "  /memory    # inspect project memory policy when needed"
 fi
 echo
 echo "Daily flow after setup:"
 echo "  herdr    # optional"
 echo "  cd <project>"
 echo "  pi"
-echo "  /task Implement <task>  # parent may auto-delegate scout/planner/reviewer"
+echo "  /workflow task Implement <task>  # parent may auto-delegate scout/planner/reviewer"

package/scripts/team-doctor.sh

@@ -149,8 +149,6 @@ if (!runtime.teamRolloutReady) {
 for (const rel of [
   "package.json",
   "packages/piagent-core/extensions/piagent-guard.ts",
-  "packages/piagent-core/prompts/onboard-project.md",
-  "packages/piagent-core/prompts/memory-policy.md",
   "packages/piagent-core/prompts/platform-improve.md",
   "packages/piagent-core/prompts/be-to-fe.md",
   "packages/piagent-core/prompts/task.md",
@@ -374,10 +372,10 @@ const projectContextPath = path.join(projectPath, ".pi", "project-context.md");
 if (fs.existsSync(projectContextPath)) {
   const projectContext = fs.readFileSync(projectContextPath, "utf8");
   if (/Generated:\s*not yet/i.test(projectContext)) {
-    warnings.push("project onboarding snapshot is still pending; run /onboard-project in Pi after login/model selection");
+    warnings.push("project onboarding snapshot is still pending; run /onboard run in Pi after login/model selection");
   }
 } else {
-  warnings.push("project has no .pi/project-context.md; run setup/init or /onboard-project");
+  warnings.push("project has no .pi/project-context.md; run setup/init or /onboard run");
 }
 
 const contextIndexConfig = projectProfile && typeof projectProfile.contextIndex === "object" && !Array.isArray(projectProfile.contextIndex)
@@ -402,8 +400,8 @@ if (contextIndexEscapes) {
     const edges = Array.isArray(contextIndex.edges) ? contextIndex.edges : [];
     const citations = Array.isArray(contextIndex.citations) ? contextIndex.citations : [];
     if (contextIndex.schemaVersion !== 1) warnings.push("project context index schemaVersion should be 1");
-    if (contextIndex.summary === "Pending. Run /onboard-project after login/model selection to generate the compact project context index.") {
-      warnings.push("project context index is still pending; run /onboard-project after login/model selection");
+    if (/Pending\.\s+Run \/onboard(?: run)? after login\/model selection/i.test(String(contextIndex.summary ?? ""))) {
+      warnings.push("project context index is still pending; run /onboard run after login/model selection");
     }
     if (nodes.length === 0) warnings.push("project context index has no nodes yet");
     if (contextIndex?.policy?.requireCitations !== false && citations.length === 0) warnings.push("project context index has no citations yet");
@@ -411,13 +409,13 @@ if (contextIndexEscapes) {
     if (edges.length > (contextIndex?.policy?.maxEdges ?? 240)) warnings.push("project context index has more edges than policy maxEdges");
   }
 } else {
-  warnings.push(`project has no ${contextIndexRelativePath}; run setup/init or /onboard-project`);
+  warnings.push(`project has no ${contextIndexRelativePath}; run setup/init or /onboard run`);
 }
 
 const memorySummaryPath = path.join(projectPath, ".pi", "memory", "memory_summary.md");
 const memoryHandbookPath = path.join(projectPath, ".pi", "memory", "MEMORY.md");
 if (!fs.existsSync(memorySummaryPath) || !fs.existsSync(memoryHandbookPath)) {
-  warnings.push("project has no .pi/memory scaffold; run setup/init or use /memory-policy before relying on project memory");
+  warnings.push("project has no .pi/memory scaffold; run setup/init or use /memory before relying on project memory");
 }
 
 if (strictShare) {

package/scripts/verify-local.sh

@@ -56,10 +56,6 @@ required_files=(
   "$ROOT/packages/piagent-core/extensions/runtime-evidence.js"
   "$ROOT/packages/piagent-core/security/sensitive-data.js"
   "$ROOT/packages/piagent-core/capabilities/capability-core.js"
-  "$ROOT/packages/piagent-core/prompts/onboard-project.md"
-  "$ROOT/packages/piagent-core/prompts/piagent-commands.md"
-  "$ROOT/packages/piagent-core/prompts/model-options.md"
-  "$ROOT/packages/piagent-core/prompts/memory-policy.md"
   "$ROOT/packages/piagent-core/prompts/platform-improve.md"
   "$ROOT/packages/piagent-core/prompts/be-to-fe.md"
   "$ROOT/packages/piagent-core/prompts/scout.md"
@@ -329,10 +325,15 @@ require_documented "piagent_context_budget" "$ROOT/packages/piagent-core" "$ROOT
 require_documented "piagent_tool_policy_check" "$ROOT/packages/piagent-core" "$ROOT/docs" "$ROOT/templates/project/AGENTS.md"
 require_documented "piagent_task_gate_check" "$ROOT/packages/piagent-core" "$ROOT/docs" "$ROOT/templates/project/AGENTS.md"
 require_documented "piagent_usage_snapshot" "$ROOT/packages/piagent-core" "$ROOT/docs" "$ROOT/templates/project/AGENTS.md"
-require_documented "/piagent-usage" "$ROOT/packages/piagent-core" "$ROOT/docs" "$ROOT/README.md"
+require_documented "/usage" "$ROOT/packages/piagent-core" "$ROOT/docs" "$ROOT/README.md"
 require_documented "piagent_document_read" "$ROOT/packages/piagent-core" "$ROOT/docs"
 require_documented "additionalReadRoots" "$ROOT/docs" "$ROOT/schemas/project-profile.schema.json"
-require_documented "/piagent-commands" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/prompts"
+require_documented "/commands" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
+require_documented "/workflow" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
+require_documented "/name" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
+require_documented "/fresh" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
+require_documented "/context" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
+require_documented "/permission" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
 require_documented "auto-delegation" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/prompts" "$ROOT/templates/project/AGENTS.md"
 require_documented "Subagents used/not used" "$ROOT/packages/piagent-core/prompts" "$ROOT/docs/auto-delegation-policy.md"
 require_documented "Subagent orchestration capabilities" "$ROOT/README.md" "$ROOT/docs/subagent-orchestration-capabilities.md"
@@ -415,13 +416,13 @@ require_documented "piagent_memory_citation_record" "$ROOT/packages/piagent-core
 require_documented "piagent_context_index_status" "$ROOT/packages/piagent-core" "$ROOT/docs"
 require_documented "piagent_context_index_record" "$ROOT/packages/piagent-core" "$ROOT/docs"
 require_documented "piagent_context_index_search" "$ROOT/packages/piagent-core" "$ROOT/docs"
-require_documented "/context-index" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/prompts"
-require_documented "/onboard-project" "$ROOT/README.md" "$ROOT/docs" "$ROOT/templates/project/AGENTS.md"
-require_documented "/memory-policy" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/prompts"
-require_documented "/model-options" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/prompts"
-require_documented "anthropic/claude" "$ROOT/README.md" "$ROOT/docs/model-options.md" "$ROOT/packages/piagent-core/prompts/model-options.md"
-require_documented "gpt-5.6" "$ROOT/README.md" "$ROOT/docs/model-options.md" "$ROOT/packages/piagent-core/prompts/model-options.md"
-require_documented "claude-fable-5" "$ROOT/README.md" "$ROOT/docs/model-options.md" "$ROOT/packages/piagent-core/prompts/model-options.md"
+require_documented "/context index" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
+require_documented "/onboard" "$ROOT/README.md" "$ROOT/docs" "$ROOT/templates/project/AGENTS.md"
+require_documented "/memory" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
+require_documented "/model-options" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
+require_documented "anthropic/claude" "$ROOT/README.md" "$ROOT/docs/model-options.md" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
+require_documented "gpt-5.6" "$ROOT/README.md" "$ROOT/docs/model-options.md" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
+require_documented "claude-fable-5" "$ROOT/README.md" "$ROOT/docs/model-options.md" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
 require_documented "piagent-models" "$ROOT/README.md" "$ROOT/docs/model-options.md"
 require_documented "enabledModels" "$ROOT/templates/global/settings.json" "$ROOT/docs/model-options.md" "$ROOT/scripts/configure-model-scope.sh"
 require_documented "piagent-mcp" "$ROOT/README.md" "$ROOT/docs/mcp-and-tools.md" "$ROOT/scripts/mcp-manage.mjs"
@@ -455,9 +456,9 @@ if grep -E '@latest|"@upstash/context7-mcp"|"chrome-devtools-mcp"|"@playwright/m
   exit 1
 fi
 require_documented "Ctrl+L" "$ROOT/README.md" "$ROOT/docs/model-options.md" "$ROOT/docs/team-onboarding.md" "$ROOT/docs/quickstart-vietnamese.md"
-require_documented "/platform-improve" "$ROOT/packages/piagent-core/prompts" "$ROOT/docs"
-require_documented "/be-to-fe" "$ROOT/packages/piagent-core/prompts" "$ROOT/docs"
-require_documented "/scout" "$ROOT/packages/piagent-core/prompts" "$ROOT/docs" "$ROOT/README.md"
+require_documented "/workflow platform-improve" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
+require_documented "/workflow be-to-fe" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
+require_documented "/workflow scout" "$ROOT/README.md" "$ROOT/docs" "$ROOT/packages/piagent-core/extensions/piagent-guard.ts"
 require_documented "piagent_context_preflight" "$ROOT/packages/piagent-core" "$ROOT/docs"
 require_documented ".pi/task-inbox" "$ROOT/.gitignore" "$ROOT/templates/project/.pi/.gitignore" "$ROOT/docs"
 require_documented "Task Implementation Contract" "$ROOT/docs" "$ROOT/packages/piagent-core/prompts"
@@ -478,7 +479,7 @@ test -s "$ROOT/tests/policy-core.test.mjs"
 # the file readable.
 non_text_sources="$(
   cd "$ROOT" && git ls-files -z -- '*.js' '*.mjs' '*.ts' '*.json' '*.md' '*.sh' '*.html' '*.css' \
-    | LC_ALL=C xargs -0 -I{} sh -c 'grep -qI "" "{}" 2>/dev/null || echo "{}"'
+    | LC_ALL=C xargs -0 -n 1 sh -c '[ ! -f "$1" ] || grep -qI "" "$1" 2>/dev/null || echo "$1"' sh
 )"
 if [[ -n "$non_text_sources" ]]; then
   echo "Source files contain control bytes, so grep-based gates and diffs skip them:"

package/templates/project/.pi/context-index.json

@@ -3,7 +3,7 @@
   "projectId": "replace-me",
   "profileMode": "generic",
   "source": "manual",
-  "summary": "Pending. Run /onboard-project after login/model selection to generate the compact project context index.",
+  "summary": "Pending. Run /onboard run after login/model selection to generate the compact project context index.",
   "generatedAt": "not-configured",
   "updatedAt": "not-configured",
   "policy": {
@@ -20,6 +20,6 @@
   "edges": [],
   "citations": [],
   "warnings": [
-    "context index pending; run /onboard-project"
+    "context index pending; run /onboard run"
   ]
 }

package/templates/project/.pi/project-context.md

@@ -4,7 +4,7 @@
 
 - Generated: not yet
 - Profile: see `.pi/piagent-profile.json`
-- Model/pass: run `/onboard-project` after Pi login and model selection
+- Model/pass: run `/onboard run` after Pi login and model selection
 - Scope: pending
 
 ## Purpose
@@ -16,13 +16,13 @@ Before the first real implementation task in this project:
 1. Open Pi in the project.
 2. Run `/login` if needed.
 3. Select the intended provider/model.
-4. Run `/onboard-project`.
+4. Run `/onboard run`.
 
 The selected model should inspect the project read-only, then replace this file with a concise project map, required context, verification matrix, and high-risk areas.
 
 ## Update triggers
 
-Regenerate with `/onboard-project` when:
+Regenerate with `/onboard run` when:
 
 - source layout changes;
 - architecture or domain ownership changes;

package/templates/project/AGENTS.md

@@ -7,8 +7,8 @@ This project uses Pi Agent Platform.
 Before implementation:
 
 1. Load `.pi/piagent-profile.json` with `piagent_context`.
-2. If `.pi/project-context.md` still says "Generated: not yet", run `/onboard-project` after login/model selection before the first implementation task.
-3. Check `.pi/context-index.json` with `/context-index` or `piagent_context_index_status` when available; use it only as an advisory navigation map.
+2. If `.pi/project-context.md` still says "Generated: not yet", run `/onboard run` after login/model selection before the first implementation task.
+3. Check `.pi/context-index.json` with `/context index` or `piagent_context_index_status` when available; use it only as an advisory navigation map.
 4. Read `.pi/project-context.md` and required context files listed in the profile; use `piagent_context_budget` for large files.
 5. Check project memory with `piagent_memory_status` when relevant; memory is advisory, source files are authoritative.
 6. Start source-changing work with `piagent_task_start`.
@@ -16,10 +16,10 @@ Before implementation:
 8. Use `piagent_exec_policy_check` before high-impact or complex shell commands.
 9. Use `piagent_tool_policy_check` before non-piagent MCP/app tools.
 10. Use MCP/tools only when declared in profile.
-11. Use `piagent_usage_snapshot` or `/piagent-usage` when the user asks about token/context usage or session follow-up.
+11. Use `piagent_usage_snapshot` or `/usage` when the user asks about token/context usage or session follow-up.
 12. Record context/verify/trace with `piagent_context_record`, `piagent_verify_record`, and `piagent_trace_record`. Passing verify evidence must be the exact command from `task.verifyCommands`.
 13. Run `piagent_task_gate_check` before DONE.
-14. For unclear tasks, use `/discuss` first; do not implement while requirements are unresolved.
+14. For unclear tasks, use `/workflow discuss` first; do not implement while requirements are unresolved.
 15. For external source repos, use `piagent-source-cache` and read targeted files only.
 16. For medium/large tasks, auto-delegate independent read-only scout/planning/review work to `piagent-scout`, `piagent-planner`, `piagent-reviewer`, or `piagent-oracle` when `pi-subagents` is available. Do not require the user to type `/run` for normal task orchestration.
 17. Keep implementation single-writer by default. Use parallel writers only with explicit user approval or safe worktree isolation and disjoint write sets.

package/packages/piagent-core/prompts/memory-policy.md

@@ -1,52 +0,0 @@
-# Memory Policy
-
-Use this workflow to inspect or configure project memory for a Pi Agent Platform project.
-
-## Goals
-
-- Keep durable project knowledge across sessions without forcing the model to re-scout the whole repo.
-- Keep memory explicit, compact, inspectable, and safe.
-- Avoid storing secrets, credentials, raw private data, or large source excerpts.
-
-## Workflow
-
-1. Call `piagent_context` with `detail=full`.
-2. Call `piagent_memory_status` with `detail=full`.
-3. If memory files are missing, explain that project memory can be initialized by project init/setup or by saving the first explicit note.
-4. If the user asks to remember a stable fact, call `piagent_memory_note`.
-5. If the user asks what is remembered, call `piagent_memory_search` or inspect the memory files.
-6. If the user asks about automatic/background memory, explain the tradeoff:
-   - default platform memory is explicit-only and project-scoped;
-   - external Pi memory packages can be installed separately after source review;
-   - background extraction should be benchmarked before being enabled for teams.
-
-## Rules
-
-- Memory is advisory. Current repo files, task contracts, and user instructions are authoritative.
-- Do not rely on memory alone for implementation decisions.
-- Cite memory use visibly in the response when it materially affects a decision: `Memory cited: <file/path>`.
-- Do not hand-edit generated memory state, SQLite catalogs, snapshots, or runtime cache files.
-- Keep reusable workflow names and prompts in English.
-
-## Recommended outputs
-
-For status:
-
-```text
-Memory mode: manual
-Scope: project
-Summary: .pi/memory/memory_summary.md
-Handbook: .pi/memory/MEMORY.md
-Write policy: explicit-only
-External packages: optional, review before install
-```
-
-For a remember request:
-
-```text
-Saved memory:
-- category:
-- title:
-- path:
-- redacted:
-```

package/packages/piagent-core/prompts/model-options.md

@@ -1,68 +0,0 @@
----
-description: Explain native Pi model selector/scope options across model families
-argument-hint: "[task type, budget, or provider preference]"
----
-
-Explain Pi provider/model/thinking selector options for:
-
-```text
-$ARGUMENTS
-```
-
-Mandatory flow:
-
-1. Call `piagent_context`.
-2. Call `piagent_usage_snapshot` if the user asks about current model/context/cost.
-3. Make clear that model switching is done by Pi native UI, not by agent deciding for the user:
-   - `/model` or Ctrl+L for selector;
-   - `/scoped-models` for cycle scope;
-   - Ctrl+P / Shift+Ctrl+P for model cycling;
-   - Shift+Tab for thinking level.
-4. Include both model families — the OpenAI Codex models and the Anthropic Claude models — unless the user explicitly restricts the provider. Name them as model families, never as another tool.
-5. Do not limit the answer to one or two legacy examples. Include the current latest-family options: the Codex 5.5/5.6 models and the Claude Sonnet/Opus/Fable models, where relevant.
-6. If the user needs available model versions, recommend `piagent-models`, `/model`, or `pi --list-models`.
-7. Explain thinking levels as task effort, not as quality guarantee.
-8. Do not claim token/cost savings without benchmark evidence.
-
-Current latest-family catalog to consider:
-
-OpenAI Codex models (provider `openai-codex`):
-
-- `openai-codex/gpt-5.3-codex-spark`
-- `openai-codex/gpt-5.4-mini`
-- `openai-codex/gpt-5.4`
-- `openai-codex/gpt-5.5`
-- `openai-codex/gpt-5.6-luna`
-- `openai-codex/gpt-5.6-sol`
-- `openai-codex/gpt-5.6-terra`
-
-Anthropic Claude models (provider `anthropic`):
-
-- `anthropic/claude-haiku-4-5`
-- `anthropic/claude-sonnet-4-5`
-- `anthropic/claude-sonnet-4-6`
-- `anthropic/claude-sonnet-5`
-- `anthropic/claude-opus-4-5`
-- `anthropic/claude-opus-4-6`
-- `anthropic/claude-opus-4-7`
-- `anthropic/claude-opus-4-8`
-- `anthropic/claude-fable-5`
-
-Default selector scope seeded by setup:
-
-| Preset | OpenAI Codex model | Anthropic Claude model | Use when |
-|---|---|---|---|
-| Fast scout | `openai-codex/gpt-5.4-mini:low` | `anthropic/claude-haiku-4-5:low` | quick reads, simple Q&A |
-| Balanced implement | `openai-codex/gpt-5.5:medium` | `anthropic/claude-sonnet-5:medium` | normal source tasks |
-| Hard implement | `openai-codex/gpt-5.6-luna:xhigh` or `openai-codex/gpt-5.5:xhigh` | `anthropic/claude-sonnet-5:xhigh` | multi-file/debug/contract mapping |
-| Strategic/deep | `openai-codex/gpt-5.6-sol:xhigh` | `anthropic/claude-opus-4-7:max` or `anthropic/claude-opus-4-8:max` | architecture, large refactor, high-risk review |
-| Huge-context scout | `openai-codex/gpt-5.6-terra:xhigh` | `anthropic/claude-fable-5:max` | large repo/docs synthesis |
-
-Output:
-
-- Current model/context if available.
-- How to open/select/change model in Pi UI.
-- What `enabledModels`/scope contains.
-- Codex model family options.
-- Claude model family options.
-- How to inspect/re-apply model scope from terminal.