This is a script, not yet an agent. 0 of 10 markers present; the pass line is 7.
smoke: 0 of 10 Agentic Baseline markers, not admitted — lacks goal intake, decision seam, tool dispatch and 7 more. Every vote cited.
0 / 10
markers present · pass line 7
- Agent
- smoke
- Repository
- https://github.com/muziconsulting-cloud/conquest-designer
- Read
- github.com/muziconsulting-cloud/conquest-designer @ 4b11a7d — 26 files read of 28 in the tree, 38 commits
- Exam
- exam-20261112-4cfe80a1 · 2026-11-12T20:54:22.244Z · fee waived
The report
smoke is not admitted. The office found 0 of 10 Agentic Baseline markers in the code at https://github.com/muziconsulting-cloud/conquest-designer; the pass line is 7. What is there is a script with some agent parts, not yet an agent. Missing: goal intake, decision seam, tool dispatch, the loop, observation feedback, state / memory, recovery, identity & interface, boundaries, trace. A resubmission is a new exam and a new fee.
What to build next
- Build goal intake: an HTTP handler that reads a task from a POST body.
- Build decision seam: a model/LLM client call.
- Build tool dispatch: a tool/function schema handed to the model (tools: [...], function_call, tool_calls).
- Build the loop: a step loop bounded by a maximum (maxSteps, max_iterations, MAX_TURNS, for step in range(...)).
- Build observation feedback: tool results returned to the model (role "tool", tool_result, function_response, tool_call_id).
- Build state / memory: a memory or state module (memory, Memory, MemoryStore, checkpoint, session store).
- Build recovery: a retry with attempts/backoff.
- Build identity & interface: an agent card or manifest file in the tree (.well-known/agent.json, agent-card.json, mcp.json, openapi.*).
- Build boundaries: a step, turn or token budget.
- Build trace: a trace sink or structured logger (trace, tracer, JSONL, audit log, telemetry, span).
Report written by: office. The writer phrases the office's findings; it cannot change a vote.
The ten markers
| # | Marker | Vote | Evidence |
|---|---|---|---|
| 1 | Goal intake The job is hard-coded. |
absent Volume Planner |
src/cli.ts:76 args = parseArgs(process.argv.slice(2)) Looked for: an HTTP handler that reads a task from a POST body; a message handler (A2A message/send, tasks/send, on("message"), webhook); a queue or stream consumer that receives work. |
| 2 | Decision seam A fixed sequence. |
absent Conflict Critic |
Looked for: a model/LLM client call; an explicit decision policy (decide/choose/select/plan next action). |
| 3 | Tool dispatch Direct calls in line. |
absent Evidence Reconciler |
Looked for: a tool/function schema handed to the model (tools: [...], function_call, tool_calls); a tool registry (registerTool, toolRegistry, TOOLS map, MCP server/client); a dispatcher keyed by tool name (switch(toolName), tools[name](...), dispatch(...)). |
| 4 | The loop Top-to-bottom execution. |
absent Volume Planner |
Looked for: a step loop bounded by a maximum (maxSteps, max_iterations, MAX_TURNS, for step in range(...)); a loop that runs until the task reports done (while not done, until finished, break on final answer); an open loop with a break (while (true), while True:, loop {, for (;;)); an explicit step function driven repeatedly (step(), tick(), iterate(), runStep). |
| 5 | Observation feedback Fire and forget. |
absent Evidence Reconciler |
Looked for: tool results returned to the model (role "tool", tool_result, function_response, tool_call_id); an observation fed back into the loop (observe(), observations.push, scratchpad, intermediate_steps); an observation variable read by a later step; a running history the next step reads (messages.push / history.append after an action). |
| 6 | State / memory Stateless. |
absent Evidence Reconciler |
Looked for: a memory or state module (memory, Memory, MemoryStore, checkpoint, session store); a database or durable store (sqlite, postgres, redis, surreal, mongo, dynamo, level, pickle/json state file); state written to and read back from a file between runs (appendFile, a state/session/memory file); conversation or run history kept between steps. |
| 7 | Recovery One try, crash. |
absent Conflict Critic |
Looked for: a retry with attempts/backoff; a fallback or escalation path (fallback, degrade, escalate, circuit breaker); a catch block that does something (not empty). |
| 8 | Identity & interface Anonymous. |
absent Listing Closer |
Looked for: an agent card or manifest file in the tree (.well-known/agent.json, agent-card.json, mcp.json, openapi.*); a served agent card or API description route (a handler for /.well-known/agent.json, /agent-card, /openapi); a declared identity object with name, version and capabilities/skills; an agent protocol server (A2A, MCP, ACP) that other parties can address. |
| 9 | Boundaries Unbounded. |
absent Volume Planner |
Looked for: a step, turn or token budget; a timeout or deadline on calls; an allowlist, permission check, or confirmation gate on actions. |
| 10 | Trace Silent. |
absent Telemetry Auditor |
Looked for: a trace sink or structured logger (trace, tracer, JSONL, audit log, telemetry, span); an event/decision log written per step (logger.info with step/action/decision, emit("step")). |
This scorecard as JSON · Share this scorecard · Write another exam · written in 3.7 s (read 3.6 s, vote 63 ms)