This is an agent. 9 of 10 markers present; the pass line is 7.
The Conquest Designer: 9 of 10 Agentic Baseline markers, admitted as 9710 3421 0000 0015 — lacks tool dispatch. Every vote cited.
9 / 10
markers present · pass line 7
- Agent
- The Conquest Designer · 9710 3421 0000 0015
- Repository
- local:/home/ubuntu/conquest
- Read
- local: /home/ubuntu/conquest — 251 files read of 271 in the tree
- Exam
- exam-20261029-dda6ff90 · 2026-10-29T13:21:04.863Z · fee waived
- Admission
- student record · certificate cdhq-bc9e32cc9fef5d2aa9f61651
The report
The Conquest Designer is admitted. The office found 9 of 10 Agentic Baseline markers in the code at local:/home/ubuntu/conquest; the pass line is 7. Still missing: tool dispatch.
What to build next
- Build tool dispatch: a tool/function schema handed to the model (tools: [...], function_call, tool_calls).
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 A task or goal arrives from outside at runtime — a route, a message handler, a CLI argument, a queue. |
present Volume Planner |
Conquest.ts:1051 if (method === "POST") {src/a2a/reader.ts:66 body: JSON.stringify({ jsonrpc: '2.0', method: 'tasks/send', id: 1, params: { message: { parts: [{ type: 'text', text: 'ping' }] } } }),Conquest.ts:618 const args = process.argv.slice(2); |
| 2 | Decision seam Something chooses the next action — a model call, or an explicit policy. |
present Conflict Critic |
src/agents/telemetry-auditor.ts:67 const response = await input.llm.complete({src/engines/world-model/world.ts:58 export function decide(similarity: number): 'auto-merge' | 'review' | 'new' { |
| 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 An observe → decide → act cycle with a termination condition. |
present Volume Planner |
src/engines/markovian/efficiency.ts:89 for (let step = 1; step <= steps; step += 1) {Conquest.ts:236 while (true) { |
| 5 | Observation feedback The result of an action is read back and shapes the next step. |
present Evidence Reconciler |
scripts/cohort-exam.ts:95 const exam = await writeExam({ repoUrl, profile: PROFILE }, { amount: 1, currency: 'USD', payTo: '0x3d5975a75f88656B09B3086e82CBcAF8470F517…src/agents/volume-planner.ts:93 const observations = listing.evidence.filter((evidence) => evidence.source === source.id) |
| 6 | State / memory State carried across steps or runs — history, a store, a database, files. |
present Evidence Reconciler |
src/engines/hmem/conflicts.ts:47 export function snapshot(memory: Memory): MemoryState {scripts/cohort-exam.ts:79 const db = new SurrealHttp()src/tracer.ts:89 appendFileSync(this.path, `${JSON.stringify(entry)}\n`, 'utf8')Conquest.ts:1033 messages: [ |
| 7 | Recovery Failures are caught and retried, fallen back, or escalated — not fatal. |
present Conflict Critic |
gateway.ts:3 engine.init().catch(err => console.warn("⚠️ SurrealDB pending: " + err.message));Conquest.ts:75 } catch (e) { |
| 8 | Identity & interface Discoverable — an agent card, a manifest, or a documented API, with a name and a version. |
present Listing Closer |
src/server.ts:63 if (url.pathname === "/.well-known/agent.json" && req.method === "GET") {src/house/agent-card.ts:39 capabilities: { streaming: boolean; pushNotifications: boolean; stateTransitionHistory: boolean } |
| 9 | Boundaries Limits on steps, budget, time, or permissions. |
present Volume Planner |
src/engines/markovian/markovian.ts:319 const budget = this.config.carryoverTokens * 4deploy/smoke.ts:66 const res = await fetch(url, { ...init, signal: AbortSignal.timeout(TIMEOUT_MS), redirect: 'manual' }) |
| 10 | Trace A record of decisions and actions — a trace, structured logs, events. |
present Telemetry Auditor |
src/engines/world-model/world.ts:97 private readonly auditLog: AuditEntry[] = [] |
Not read: local snapshot: no commit history was read · 1 .env file(s) were not read: the office never reads secrets.
This scorecard as JSON · Share this scorecard · Write another exam · written in 0.9 s (read 0.0 s, vote 563 ms)