Naming Decision Packet

This packet records the current public naming recommendation for LoopX. It is intentionally lightweight: the project is still young, so the goal is to avoid a premature repository rename while giving maintainers a crisp brand, category, and tagline to test in first-contact conversations.

Recommendation

Keep the repository and product name:

LoopX

Use this hero promise near the first screen:

Always-on agent teams, governed by human judgment

Use this control-plane category line next to it:

Gate-aware human-in-the-loop control plane

Use this supporting category/tagline language when more explanation is needed:

dynamic goal control plane

The public story should be:

LoopX is the dynamic goal control plane around agent loops. Codex, Claude Code, Cursor, terminal agents, and benchmark runners execute bounded work; LoopX turns a static goal into long-running state: gates, todos, claims, scopes, quota, evidence, and handoffs stay visible across those loops.

Do not rename the repo to lifetime-loopx or dynamic-loopx now. Use "dynamic goal control plane" as category language until there is stronger evidence that users remember another name better than LoopX.

Why Not Rename Yet

LoopX has two advantages:

• it is short enough to say in conversation and fit in command names;
• it keeps the current CLI, docs, repo, package, and existing public links stable while the product surface is still moving.

The name is not perfectly distinctive. "Goal" can sound like a task-list or Codex goal-mode feature, and "Harness" needs the first screen to explain the control-plane role. But this can be handled with the category and comparison copy:

Not an agent runtime; the control plane around the runtime.
Not Codex goal mode; the dynamic goal state that survives across executor
loops.

A repo rename would be noisy now: it would create link churn before the project has enough public proof that a new name improves recall, search, and comprehension.

Candidate Comparison

Candidate	Strength	Weakness	Decision
LoopX	Short, current, command-friendly, broad enough for engineering and creator/operator cases.	Needs category copy to avoid confusion with Codex goal mode or todo apps.	Keep as brand.
Lifetime LoopX	Makes the long-horizon ambition explicit.	Long, heavy, awkward as a repo/package name, and can sound abstract before users understand the product.	Use as explanatory language, not brand.
Dynamic LoopX	Captures the contrast with static goal prompts and one-shot automation.	Still long as a repo/package name, and "dynamic" can sound generic without the control-plane proof.	Use as category signal, not brand.
Goal Control Plane	Directly names the category.	Too generic; weak brand; less distinctive in search or conversation.	Use as occasional explanation only.
Agent Control Plane	Clear for agent-infrastructure audiences.	Overclaims the scope and implies LoopX is a runtime/orchestrator.	Avoid as primary framing.
Long-Horizon Control Plane	Emphasizes the real problem.	Broad and technical; less connected to the user's concrete goal.	Use in long-form docs when useful.
GoalOS / AgentOS style names	Memorable and product-like.	Overclaims platform scope and invites runtime expectations.	Avoid for now.

Naming Architecture

Use three layers:

1. Brand: LoopX
2. Hero promise: Always-on agent teams, governed by human judgment
3. Category: Gate-aware human-in-the-loop control plane
4. Support phrase: dynamic goal control plane

This lets the project keep a stable name while still sharpening the product category.

Example first-screen stack:

LoopX

Always-on agent teams, governed by human judgment.

Gate-aware human-in-the-loop control plane.

Dynamic goal control plane for long-running agents.

Chinese-first stack:

LoopX

Always-on agent teams, governed by human judgment
Gate-aware human-in-the-loop control plane
Dynamic goal control plane for long-running agents
让多个 agent 昼夜接力，把人的判断留在控制面。

LoopX 不是替代 Codex goal/automation，而是给这些 executor loop
提供动态长期目标控制面。

When To Use Each Phrase

Use LoopX when naming the product, repository, CLI, examples, and showcase cases.

Use Always-on agent teams, governed by human judgment when the first-contact surface needs a stronger product promise than "state management" without overclaiming autonomous production control.

Use Gate-aware human-in-the-loop control plane when a first-contact reader needs the technical category immediately.

Use dynamic goal control plane when explaining why this is larger than one agent session, one terminal command, or one benchmark run. Use lifetime-goal only as a secondary explanation for time horizon, not as the main category.

Use Your agents keep the night shift. You keep the judgment. as a social or demo tagline, not as a formal architecture term.

Avoid agent teammate, agent framework, autonomous platform, or agent OS unless a future product surface actually owns those capabilities.

Validation Signal Before Any Rename

Do not rename from LoopX unless at least one lightweight validation cycle shows the current brand is the bottleneck.

A good validation cycle:

1. Show five new users the README or launch post for less than one minute.
2. Ask them to answer three questions without hints:
3. Is LoopX replacing Codex/Claude/Cursor, or wrapping them?
4. What problem does it solve in long-running agent work?
5. What phrase do they remember after closing the page?
6. Compare recall of LoopX, dynamic goal control plane, lifetime-goal control plane, and human-in-the-loop control plane.
7. Record only public-safe aggregated notes: no raw chat transcripts, personal names, screenshots, internal links, or private project context.
8. Rename only if users consistently remember another name better and also understand the runtime-vs-control-plane distinction.

Until that happens, improve first-screen copy, showcases, and diagrams before changing the brand.

Practical Copy Rules

• Pair the brand with the category on first mention: LoopX, an always-on control plane for agent teams governed by human judgment.
• When users confuse it with Codex goal mode, answer with the executor-loop split: Codex does bounded work; LoopX preserves dynamic goal state.
• Use "dynamic goal" for the category; use "lifetime-goal" in explanatory paragraphs only when the time horizon needs emphasis.
• Keep bilingual hero copy for Chinese-first surfaces, but keep English README body primarily English.
• Treat naming claims as outreach language. Do not add brittle smoke tests that assert exact marketing copy unless the wording becomes a stable public contract.

Current Decision

Decision: keep LoopX.

Rationale: the name is serviceable, current, and compact; the sharper work is to make the category unmistakable through first-screen copy, showcases, and the control-plane board.

Next review trigger: revisit naming only after first-contact validation shows that the name, not the explanation surface, is the main comprehension blocker.