RunGlance MCP server

StatusGlance

StatusGlance is Openly Useful's visibility family: StatusGlance / Readiness turns one repository-local manifest into evidence-backed delivery readiness; StatusGlance / RunGlance shows observed run activity without model polling. Both are portable across Claude, OpenAI/Codex, and generic CLIs, with optional bounded read-only MCP entrypoints.

This compatibility-first branding release retains the project-status and runglance plugin IDs, npm/MCP names, commands, storage paths, repository URL, and Apache-2.0 license. Existing integrations do not need a rename migration. A status request reads existing evidence once; it does not start work, verification, monitoring, or installation.

The delivery loop

Conductor Swarm owns execution: observe changes → shape coherent work → run ready independent lanes in parallel → verify and integrate at milestones → adapt → continue authorized work or finish. Pickup repairs missing context; Continuity handles actual transfers. These are complementary tools, not mandatory extra agents or serialized stages. Agent Workflow Swarms defines the execution framework.

StatusGlance observes that loop. Use existing host events, task IDs, and milestone evidence. No event stream means unknown activity. A local setup file does not prove host installation or connection. Repeated revisions call for an approach change within authorized execution, not an ever-growing review loop. Operations Pulse remains the opt-in recurring operations sibling; Skill Feedback Engine proposes durable improvements only when requested.

Alpha, Beta, and Live readiness

The optional versioned delivery extension references existing manifest tasks. Ordered stage scopes accumulate accepted work once; each shows its own denominator, remaining tasks and milestones, next integrated milestone, and acceptance gaps. Empty scope remains unknown. Current stage acceptance requires qualifying evidence and cleared gates; implementation progress and historical weighted credit remain separate.

The bundled dashboard is a historical demonstration, not evidence of any user's current product readiness. Configure and validate real product scope before claiming Alpha, Beta, or Live completion.

This repository contains eight related deliverables:

• the canonical portable skill at skill/project-status;
• the standalone RunGlance skill at skill/runglance;
• the dependency-free RunGlance runtime, host adapters, HUD presets, verification runner, and final receipts;
• optional read-only Project Status and RunGlance MCP entrypoints under packages/mcp;
• generated OpenAI/Codex and Claude plugin wrappers under plugins/;
• deterministic Project Status and RunGlance ZIP distributions produced by scripts/package-skill.mjs and scripts/package-runglance.mjs;
• the React/Vite status prototype and its OpenAI Sites worker build.

Trust and safety model

Readiness, evidence validity, live health, and delivery activity remain separate signals. Browser refreshes do not count as monitoring. Read-only inspection is the default; manifest edits, scheduling, installation, publishing, deployment, commits, and remote writes require separate explicit actions.

Activity also separates context-window capacity, provider quota, task budget, monetary cost, progress, heartbeat freshness, and explicit lock ownership. Missing host data stays unknown; it is never displayed as zero or guessed from another metric. Passing verification proves the check and task result recorded by the receipt, not overall project readiness.

Release tooling stays repository-local. It does not install a plugin, change a personal marketplace, publish an archive, create a release, deploy Sites, or write outside its selected artifact directory.

Canonical skill and generated wrappers

Shortened. The full README is on GitHub.