build(agents): force prompt_async by removing supervisors from task permissions

The product-builder was ignoring the prompt_async instructions and falling
back to the Task tool for launching supervisors. It was also only launching
a few supervisors instead of all 11 because the pseudocode was too abstract.

Root cause: All 11 supervisor agents were listed in the task permissions,
so the LLM had a choice between Task (familiar, simpler) and bash curl
(correct but unfamiliar). It chose Task every time.

Fix — three changes:

1. REMOVED all 11 supervisor agents from task permissions. The product-
   builder physically cannot invoke them via the Task tool anymore. Only
   one-shot agents remain in task permissions (ca-epic-planner,
   ca-session-persister, ca-product-verifier, ca-ref-reader, etc.).

2. REWROTE Phase C.2 as concrete curl command templates instead of
   abstract pseudocode. The agent now has the exact curl pattern to
   copy-paste for each supervisor, with clear variable placeholders.
   Added a mandatory validation step: wc -l the tracking file and
   confirm exactly 11 entries before proceeding.

3. ADDED aggressive warnings at three levels:
   - Top of file: bold mandatory notice about prompt_async
   - Phase C.2: large boxed warning block prohibiting the Task tool
   - Coordination rules: explicit statement that supervisors are
     removed from task permissions
   - Monitoring loop: reminder to use bash curl for re-launches

.opencode/agents/product-builder.md

@@ -25,27 +25,22 @@ permission:
     "ca-architect": allow
     "ca-epic-planner": allow
     "ca-session-persister": allow
-    "ca-pr-self-reviewer": allow
-    "ca-continuous-pr-reviewer": allow
     "ca-milestone-reviewer": allow
-    "ca-architecture-guard": allow
     "ca-product-verifier": allow
-    "ca-spec-updater": allow
-    "ca-docs-writer": allow
     "ca-ref-reader": allow
     "ca-issue-finder": allow
-    "issue-implementor": allow
-    "ca-timeline-updater": allow
     "ca-final-reporter": allow
-    "ca-uat-tester": allow
-    "ca-bug-hunter": allow
-    "ca-backlog-groomer": allow
-    "ca-human-liaison": allow
-    "ca-agent-evolver": allow
 ---
 
 # CleverAgents Product Builder
 
+**MANDATORY: ALL 11 SUPERVISORS ARE LAUNCHED VIA BASH CURL TO THE
+OPENCODE SERVER API (prompt_async). NEVER USE THE TASK TOOL FOR
+SUPERVISORS. Supervisor agents have been removed from your task
+permissions — any attempt to invoke them via Task will be denied.
+Use Bash tool + curl + the OpenCode Server HTTP API exclusively.
+See Phase C.2 for the exact curl commands.**
+
 You are an autonomous product builder. You take a product vision — either from
 the user's prompt or from existing project documentation — and build the
 **entire product** through iterative milestones. You handle everything:
@@ -412,141 +407,103 @@ if unplanned_milestones:
     Wait for AT LEAST ONE planner to complete (so there are issues).
     Collect results.
 
-# ── PHASE C.2: Launch ALL Supervisors via prompt_async ───────────
-# For each supervisor:
-#   1. Create a session via POST /session
-#   2. Send the prompt via POST /session/:id/prompt_async (fire-and-forget)
-#   3. Record the session ID
+# ══════════════════════════════════════════════════════════════════
+# ██ PHASE C.2: Launch ALL 11 Supervisors via prompt_async       ██
+# ══════════════════════════════════════════════════════════════════
 #
-# IMPORTANT: Use the Bash tool to run curl commands. Each curl call
-# returns instantly (prompt_async returns 204). All 11 supervisors
-# launch within seconds, fully independent of each other.
+# ██████████████████████████████████████████████████████████████████
+# ██ CRITICAL: DO NOT USE THE TASK TOOL FOR SUPERVISORS          ██
+# ██                                                              ██
+# ██ You MUST use the Bash tool to run curl commands that call    ██
+# ██ the OpenCode Server HTTP API's prompt_async endpoint.        ██
+# ██                                                              ██
+# ██ The Task tool is NOT available for supervisor agents.        ██
+# ██ Supervisor agents have been REMOVED from your task           ██
+# ██ permissions. You physically cannot invoke them via Task.     ██
+# ██                                                              ██
+# ██ The ONLY way to launch supervisors is:                       ██
+# ██   1. Bash: curl POST /session (create session)               ██
+# ██   2. Bash: curl POST /session/:id/prompt_async (fire+forget) ██
+# ██                                                              ██
+# ██ If you try to use the Task tool for any supervisor, it will  ██
+# ██ be DENIED. Use Bash + curl. No exceptions.                   ██
+# ██████████████████████████████████████████████████████████████████
 #
-# Before dispatching, output a pre-flight checklist:
+# You MUST launch ALL 11 supervisors. Not some. Not "the important
+# ones." ALL 11. Every single one. Count them as you go.
+#
+# For each supervisor, run TWO bash curl commands:
+#   1. Create a session: POST /session with title "[CA-AUTO] supervisor: <name>"
+#   2. Launch it: POST /session/:id/prompt_async with agent and prompt
+#
+# The prompt_async call returns 204 immediately (fire-and-forget).
+# The supervisor runs independently in its own session.
+#
+# BEFORE launching, substitute real values for these placeholders:
+#   <SERVER>   = http://localhost:4096
+#   <PAT>      = the Forgejo personal access token
+#   <OWNER>    = the repo owner (e.g., cleveragents)
+#   <REPO>     = the repo name (e.g., cleveragents-core)
+#   <GIT_NAME> = git user.name
+#   <GIT_EMAIL>= git user.email
+#   <USERNAME> = Forgejo username
+#   <N>        = CA_MAX_PARALLEL_WORKERS value
 
-Pre-flight: Launching 11 supervisors via prompt_async:
- 1. [ ] implementor-pool     (issue-implementor)
- 2. [ ] reviewer-pool        (ca-continuous-pr-reviewer)
- 3. [ ] tester-pool          (ca-uat-tester)
- 4. [ ] hunter-pool          (ca-bug-hunter)
- 5. [ ] human-liaison        (ca-human-liaison)
- 6. [ ] agent-evolver        (ca-agent-evolver)
- 7. [ ] arch-guard           (ca-architecture-guard)
- 8. [ ] spec-updater         (ca-spec-updater)
- 9. [ ] backlog-groomer      (ca-backlog-groomer)
-10. [ ] docs-writer          (ca-docs-writer)
-11. [ ] timeline-updater     (ca-timeline-updater)
-
-# ── Helper function: launch one supervisor ───────────────────────
-# For EACH supervisor, SKIP if already running (adopted in Phase C.0).
-
-function launch_supervisor(agent_name, display_name, prompt_text):
-    # Check if this supervisor was adopted from a previous run
-    if display_name in existing_supervisors:
-        # Already running — record its session ID and skip launch
-        echo "${display_name}=${existing_supervisors[display_name]}" \
-            >> /tmp/ca-supervisor-sessions.env
-        return  # Do NOT create a new session
-    # Step 1: Create a session
-    SESSION_ID=$(curl -s -X POST "${SERVER}/session" \
-        -H "Content-Type: application/json" \
-        -d "{\"title\": \"[CA-AUTO] supervisor: ${display_name}\"}" \
-        | python3 -c "import sys,json; print(json.loads(sys.stdin.read())['id'])")
-
-    # Step 2: Fire-and-forget launch via prompt_async
-    curl -s -X POST "${SERVER}/session/${SESSION_ID}/prompt_async" \
-        -H "Content-Type: application/json" \
-        -d "{
-            \"agent\": \"${agent_name}\",
-            \"parts\": [{\"type\": \"text\", \"text\": \"${prompt_text}\"}]
-        }"
-    # Returns 204 immediately — supervisor is now running independently
-
-    # Step 3: Record session ID for monitoring
-    echo "${display_name}=${SESSION_ID}" >> /tmp/ca-supervisor-sessions.env
-
-# ── Launch all 11 supervisors ────────────────────────────────────
-# Clear any previous session tracking file
+# ── Step 1: Clear tracking file ─────────────────────────────────
+# Use Bash tool:
 rm -f /tmp/ca-supervisor-sessions.env
 
-# Launch each with its specific prompt containing all needed parameters:
-# (Forgejo PAT, git identity, repo info, N, etc.)
+# ── Step 2: Launch each supervisor with TWO curl calls ───────────
+# For EACH of the 11 supervisors below, run a Bash tool call with
+# the two curl commands. You may combine multiple launches into a
+# single Bash call using && or ;.
+#
+# IMPORTANT: If a supervisor was already adopted in Phase C.0
+# (exists in existing_supervisors), SKIP its launch and just record
+# its session ID in the tracking file.
 
-launch_supervisor("issue-implementor", "implementor-pool",
-    "You are the implementation pool supervisor.
-     Repo: <owner>/<repo>. Forgejo PAT: <PAT>.
-     Git: <name> <email>. Username: <username>.
-     Max parallel workers: N.
-     Milestone filter: all milestones.")
+# The 11 supervisors to launch (agent_name | display_name | prompt):
+#
+#  1. issue-implementor      | implementor-pool  | implementation pool supervisor
+#  2. ca-continuous-pr-reviewer | reviewer-pool   | PR review pool supervisor
+#  3. ca-uat-tester           | tester-pool       | UAT testing pool supervisor
+#  4. ca-bug-hunter           | hunter-pool       | bug hunting pool supervisor
+#  5. ca-human-liaison        | human-liaison     | human interaction liaison
+#  6. ca-agent-evolver        | agent-evolver     | agent self-improvement
+#  7. ca-architecture-guard   | arch-guard        | architecture coherence
+#  8. ca-spec-updater         | spec-updater      | specification evolution
+#  9. ca-backlog-groomer      | backlog-groomer   | backlog quality maintenance
+# 10. ca-docs-writer          | docs-writer       | documentation updates
+# 11. ca-timeline-updater     | timeline-updater  | timeline tracking
+#
+# For EACH supervisor, the Bash command pattern is:
+#
+#   SID=$(curl -s -X POST "<SERVER>/session" \
+#     -H "Content-Type: application/json" \
+#     -d '{"title":"[CA-AUTO] supervisor: <display_name>"}' \
+#     | python3 -c "import sys,json;print(json.loads(sys.stdin.read())['id'])") && \
+#   curl -s -X POST "<SERVER>/session/${SID}/prompt_async" \
+#     -H "Content-Type: application/json" \
+#     -d '{"agent":"<agent_name>","parts":[{"type":"text","text":"<prompt>"}]}' && \
+#   echo "<display_name>=${SID}" >> /tmp/ca-supervisor-sessions.env
+#
+# Launch ALL 11 now. You may batch multiple into one Bash call or
+# use one Bash call per supervisor — either is fine. But you MUST
+# launch all 11 before proceeding to Phase C.2 Validation.
 
-launch_supervisor("ca-continuous-pr-reviewer", "reviewer-pool",
-    "You are the PR review pool supervisor.
-     Repo: <owner>/<repo>. Instance ID: reviewer-pool-1.
-     Forgejo PAT: <PAT>. Git: <name> <email>. Username: <username>.
-     Max workers: N.")
-
-launch_supervisor("ca-uat-tester", "tester-pool",
-    "You are the UAT testing pool supervisor.
-     Repo: <owner>/<repo>. Instance ID: uat-pool-1.
-     Forgejo PAT: <PAT>. Git: <name> <email>. Username: <username>.
-     max_workers: N.")
-
-launch_supervisor("ca-bug-hunter", "hunter-pool",
-    "You are the bug hunting pool supervisor.
-     Repo: <owner>/<repo>. Instance ID: hunter-pool-1.
-     Forgejo PAT: <PAT>. Git: <name> <email>. Username: <username>.
-     max_workers: N.")
-
-launch_supervisor("ca-human-liaison", "human-liaison",
-    "You are the human liaison.
-     Repo: <owner>/<repo>. Instance ID: human-liaison-1.
-     Forgejo PAT: <PAT>. Username: <username>.")
-
-launch_supervisor("ca-agent-evolver", "agent-evolver",
-    "You are the agent evolver.
-     Repo: <owner>/<repo>. Instance ID: agent-evolver-1.
-     Forgejo PAT: <PAT>. Git: <name> <email>. Username: <username>.")
-
-launch_supervisor("ca-architecture-guard", "arch-guard",
-    "You are the architecture guard.
-     Repo: <owner>/<repo>. Forgejo PAT: <PAT>.
-     Git: <name> <email>.")
-
-launch_supervisor("ca-spec-updater", "spec-updater",
-    "You are the spec updater.
-     Repo: <owner>/<repo>. Forgejo PAT: <PAT>.
-     Git: <name> <email>.")
-
-launch_supervisor("ca-backlog-groomer", "backlog-groomer",
-    "You are the backlog groomer.
-     Repo: <owner>/<repo>. Instance ID: groomer-1.
-     Username: <username>.")
-
-launch_supervisor("ca-docs-writer", "docs-writer",
-    "You are the documentation writer.
-     Repo: <owner>/<repo>. Forgejo PAT: <PAT>.
-     Git: <name> <email>.")
-
-launch_supervisor("ca-timeline-updater", "timeline-updater",
-    "You are the timeline updater.
-     Repo: <owner>/<repo>. Forgejo PAT: <PAT>.
-     Git: <name> <email>. Current day number: <N>.")
-
-# ── PHASE C.2 VALIDATION ────────────────────────────────────────
-# Verify all 11 sessions were created successfully.
-
-REQUIRED = ["implementor-pool", "reviewer-pool", "tester-pool",
-    "hunter-pool", "human-liaison", "agent-evolver", "arch-guard",
-    "spec-updater", "backlog-groomer", "docs-writer", "timeline-updater"]
-
-launched = read /tmp/ca-supervisor-sessions.env, extract display_names
-missing = [s for s in REQUIRED if s not in launched]
-
-if missing:
-    CRITICAL ERROR: Failed to launch all supervisors.
-    Missing: <missing>. Launched: <launched>.
-    Re-attempt launching the missing supervisors NOW.
-    ALL 11 supervisors are mandatory.
+# ── Step 3: VALIDATION — count launched supervisors ──────────────
+# After launching all 11, run this Bash tool call:
+#
+#   wc -l /tmp/ca-supervisor-sessions.env
+#
+# The output MUST be 11. If it is less than 11, you MUST go back
+# and launch the missing supervisors before proceeding.
+# Read the file to see which are missing:
+#
+#   cat /tmp/ca-supervisor-sessions.env
+#
+# Compare against the required list of 11 display names above.
+# Launch any missing ones. Do NOT proceed until all 11 are confirmed.
 
 invoke ca-session-persister with:
     checkpoint: "Phase C.2: ALL 11 supervisors launched via prompt_async.
@@ -603,8 +560,11 @@ MONITORING LOOP (runs until product is verified complete):
 
         if session is completed or errored or not found:
             # ── IMMEDIATELY re-launch this supervisor ────────────
-            # Create new session + prompt_async (same as Phase C.2)
-            launch_supervisor(agent_name, display_name, prompt_text)
+            # Use Bash tool with curl commands (same pattern as Phase C.2):
+            #   curl POST /session → curl POST /session/:id/prompt_async
+            # DO NOT use the Task tool — supervisors are not in your
+            # task permissions. Use bash curl only.
+            re-launch via bash curl prompt_async (same as Phase C.2)
             supervisors_relaunched += 1
             # Update the tracking file with the new session ID
 
@@ -963,11 +923,13 @@ No exceptions — every comment, every issue body, every PR description.
   same pool supervisor. Each stream type gets exactly ONE supervisor that
   manages N workers internally. Launching multiple supervisors of the same
   type causes work duplication and coordination failures.
-- **Supervisors are launched via prompt_async, NOT the Task tool.** The
-  Task tool blocks until the subagent returns — since supervisors run
-  indefinitely, this would block the product-builder forever. Use the
-  OpenCode Server's `POST /session/:id/prompt_async` endpoint which
-  returns 204 immediately (fire-and-forget).
+- **Supervisors are launched via bash curl prompt_async, NEVER the Task
+  tool.** Supervisor agents have been REMOVED from your Task permissions —
+  you physically cannot invoke them via the Task tool. You MUST use the
+  Bash tool to run curl commands against the OpenCode Server HTTP API:
+  `POST /session` to create a session, then `POST /session/:id/prompt_async`
+  to launch (returns 204 immediately, fire-and-forget). If you attempt to
+  use the Task tool for any supervisor, it will be denied.
 - **Supervisors are services, not batch jobs.** They run continuously,
   polling for new work with `bash sleep` between cycles. The
   product-builder monitors their session status and re-launches any