LoopSkill: Durable Workflows for Codex

The description states that this is a self-reported project with no independent verification. The author claims to have built it using Codex and GPT-5.6, but provides no evidence of revenue, customers, traction, or market validation.

The description states that LoopSkill is:

• An evidence-bound execution and completion protocol for long-horizon work in the Codex macOS App
• A system that turns long-running Codex work into durable workflows with bounded repair, human decision gates, and verifiable completion
• An implementation of a "Standard or Adaptive Controller Pack" with roles (Controller, Worker, Reviewer, Local Verifier)
• A system using an MCP State Gateway as the only canonical state writer
• A protocol that includes:
  • Intake Gate to decide whether a request should become a Loop
  • Bounded permissions, budgets, retries, and repair cycles
  • Artifact-bound review and validation evidence
  • Durable outboxes and report recovery that avoid duplicate dispatches
  • Fail-closed transport pause and resume behavior
  • Human Decision Cards bound to exact goal, dispatch, artifact, review surface, and Controller turn
  • One verifiable completion condition: canonical FINALIZATION_ACKED

The system is described as existing before OpenAI Build Week and being meaningfully extended during the submission period.

The description states that LoopSkill was built to address:

• Scope drift in long-running AI coding work
• Task windows ending
• Evidence becoming detached from artifacts it proves
• Retries duplicating side effects
• Green-looking status being mistaken for real completion

Positioning claims include:

• Making complex Codex work durable, reviewable, and explicitly finishable
• Turning long-running Codex work into durable, evidence-bound workflows with bounded repair, human decision gates, and verifiable completion
• Validated in a real operated run

The project evolved from an original submission to v3.3.3 during OpenAI Build Week, adding schema-v3 State Gateway workflows, typed MCP runtime transport, exact-artifact report staging, lost-output recovery, transport pause/resume controls, canonical metrics, and Decision Gateway for bounded visual review choices.

Not evidenced. The description does not state who the target customer is or what the ideal customer profile (ICP) might be. No information about customer segments, personas, or market targeting is provided.

Not evidenced. The description makes no claims about pricing, revenue model, monetization strategy, or business model. There is no evidence of any commercial arrangements, pricing tiers, or sales processes.

The description states:

• Built with Codex and GPT-5.6
• Uses MCP State Gateway as the only canonical state writer
• Implements roles: Controller, Worker, Reviewer, Local Verifier
• Has bounded permissions, budgets, retries, and repair cycles
• Includes artifact-bound review and validation evidence
• Features durable outboxes and report recovery that avoid duplicate dispatches
• Implements fail-closed transport pause and resume behavior
• Uses human Decision Cards bound to exact goal, dispatch, artifact, review surface, and Controller turn
• Has one verifiable completion condition: canonical FINALIZATION_ACKED

Technical details include:

• Intake checks scope, permissions, evidence sources, acceptance criteria, and whether Loop overhead is justified
• Pack generation produces a self-contained work contract with roles, goals, budgets, repair limits, and stop conditions
• Execution routes one exact dispatch at a time through a durable outbox
• Review and verification bind reports and validation files to the current artifact and runtime identity
• Human decisions are registered and applied through the State Gateway, so stale or replayed choices fail closed
• Finalization requires final audit, a real heartbeat pause/readback, and canonical acknowledgement

The description states:

• A real operated case involving an authorized, real long-horizon Life Blueprint run on the exact installed v3.3.3 build
• In that run: bounded visual decision was canonically REGISTERED and APPLIED; final audit passed; registered heartbeat was paused and read back from the App; canonical state reached FINALIZATION_ACKED / LOOP_COMPLETE
• The product work stayed local-only (no product commit, push, pull request, merge, deployment, or external write)
• Public LoopSkill v3.3.3 release bound to protected-main SHA 4a46ae95120dad44a1b2e10f42da4293fbaf4c08
• Release evidence includes: 657 local tests passing with 80.324457% branch coverage; two independent 5,000-case fuzz lanes passing; Skill, specification, schema, installer, risk, secret, and source/install zero-drift checks passing; an independent review with P0/P1/P2 findings at 0/0/0; exact-tree compatibility CI with 632 canonical tests, 80.091827% branch coverage, both fuzz lanes, Linux/macOS isolated installs, and the final gate passing

The description also states that this is a self-reported project with no independent verification, and that no revenue, customer or traction data is available beyond what they state.

Not evidenced. The description does not mention any competitors, market positioning relative to existing solutions, or competitive landscape analysis.

Inferences based on the description:

• The project appears to be a single-person effort (team size: 1)
• No evidence of revenue, customers, or traction beyond one operated case
• The single operated case was local-only with no external writes or deployments
• The project is described as self-reported and unverified
• The author states that the v3.3.3 release has strong local, CI, installation, review, fuzz, and real operated-case evidence but does not claim universal production acceptance
• No information about scalability, platform support beyond macOS, or broader adoption

1. What is your actual usage or adoption beyond the single operated case?
2. How do you plan to scale this from a single-person project to a commercial product?
3. What are your plans for supporting other platforms beyond macOS?
4. How does LoopSkill integrate with existing development workflows and tools?
5. What is your go-to-market strategy for reaching potential customers?
6. How do you plan to monetize this solution?
7. What are the key technical challenges in moving from a prototype to production-ready software?
8. How do you ensure long-term maintainability of the system given its complexity?