How do I install the Meta Harness Four App Alignment skill?

Run npx skillmds@latest add grtninja/meta-harness-four-app-alignment in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Meta Harness Four App Alignment safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: makes network calls. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Meta Harness Four App Alignment?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Meta Harness Four App Alignment free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Meta Harness Four App Alignment?

grtninja (@grtninja) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Meta Harness Four App Alignment by grtninja

Name: Meta Harness Four App Alignment
Rating: 4.2 (5 reviews)
Author: grtninja

Meta Harness Four App Alignment

Align the four core local app surfaces under the shared meta-harness contract. Use when PC Control, shim/control-center, STARFRAME app, and avatar-runtime defaults or docs drift on model authority, canonical repo root, continuity, or endpoint expectations and one coordinated pass must repair them together.

grtninja Updated Sep 10, 2026

File contents

Meta-Harness Four-App Alignment

Use this skill when one pass must align the four core local app surfaces as a single contract:

PC Control governance surface
shim/control-center surface
STARFRAME app surface
avatar runtime surface

Workflow

Route the work through $skill-hub.
Capture first-party evidence before edits:
- PC Control local-agent or status-surface evidence

direct LM Studio model-lane health on :1234 and explicitly selected hosted model-lane health on :2337; :9000 is support-only
- avatar runtime health if it is expected to be live

Normalize the active contract:
- canonical repo root is G:\GitHub

http://127.0.0.1:1234/v1 is the direct LM Studio authoritative lane
http://127.0.0.1:2337/v1 is the hosted large-model authoritative lane
http://127.0.0.1:9000 is the non-model MX3 device/DFP/telemetry/RAG/support boundary

Review open diffs and contract/docs drift across the four surfaces before patching.
Repair the smallest authoritative layer first:
- governance/control contracts
- shim/router defaults
- STARFRAME app continuity/defaults
- avatar runtime defaults and health expectations
Validate each surface with targeted health, docs-lockstep, and repo-gate evidence.
Record remaining gaps and route follow-on fixes to the most specific skill lanes.

Required Evidence

PC Control evidence or status-surface output
endpoint/status results for 9000, 2337, and any expected runtime lane
exact repo/file list for each aligned surface
targeted validation output per touched surface
explicit list of any remaining degraded lanes

Route To Supporting Skills

$shim-pc-control-brain-routing for one-brain routing and port-authority fixes
$heterogeneous-stack-validation for end-to-end runtime truth checks
$docs-alignment-lock for cross-repo documentation lockstep
$skill-enforcer when boundary/policy synchronization is required
$desktop-startup-acceptance when launcher or startup behavior changed

Scope Boundary

Use this skill for coordinated four-surface contract alignment only.

Do not use it for:

isolated single-repo fixes already covered by a more specific repo skill
release-version bumping without contract/runtime review
generic workstation bring-up unrelated to the four-surface meta-harness lane

References

references/alignment-contract.md

Loopback

If one surface remains unresolved:

capture the failing contract slice with exact evidence
route the failing slice to the smallest specific skill
resume this four-app lane only after the blocking slice is repaired

---name: meta-harness-four-app-alignmentdescription: Align the four core local app surfaces under the shared meta-harness contract. Use when PC Control, shim/control-center, STARFRAME app, and avatar-runtime defaults or docs drift on model authority, canonical repo root, continuity, or endpoint expectations and one coordinated pass must repair them together.---# Meta-Harness Four-App AlignmentUse this skill when one pass must align the four core local app surfaces as a single contract:- PC Control governance surface- shim/control-center surface- STARFRAME app surface- avatar runtime surface## Workflow1. Route the work through `$skill-hub`.2. Capture first-party evidence before edits: - PC Control local-agent or status-surface evidence- direct LM Studio model-lane health on `:1234` and explicitly selected hosted model-lane health on `:2337`; `:9000` is support-only - avatar runtime health if it is expected to be live3. Normalize the active contract: - canonical repo root is `G:\GitHub`- `http://127.0.0.1:1234/v1` is the direct LM Studio authoritative lane- `http://127.0.0.1:2337/v1` is the hosted large-model authoritative lane- `http://127.0.0.1:9000` is the non-model MX3 device/DFP/telemetry/RAG/support boundary4. Review open diffs and contract/docs drift across the four surfaces before patching.5. Repair the smallest authoritative layer first: - governance/control contracts - shim/router defaults - STARFRAME app continuity/defaults - avatar runtime defaults and health expectations6. Validate each surface with targeted health, docs-lockstep, and repo-gate evidence.7. Record remaining gaps and route follow-on fixes to the most specific skill lanes.## Required Evidence- PC Control evidence or status-surface output- endpoint/status results for `9000`, `2337`, and any expected runtime lane- exact repo/file list for each aligned surface- targeted validation output per touched surface- explicit list of any remaining degraded lanes## Route To Supporting Skills- `$shim-pc-control-brain-routing` for one-brain routing and port-authority fixes- `$heterogeneous-stack-validation` for end-to-end runtime truth checks- `$docs-alignment-lock` for cross-repo documentation lockstep- `$skill-enforcer` when boundary/policy synchronization is required- `$desktop-startup-acceptance` when launcher or startup behavior changed## Scope BoundaryUse this skill for coordinated four-surface contract alignment only.Do not use it for:1. isolated single-repo fixes already covered by a more specific repo skill2. release-version bumping without contract/runtime review3. generic workstation bring-up unrelated to the four-surface meta-harness lane## References- `references/alignment-contract.md`## LoopbackIf one surface remains unresolved:1. capture the failing contract slice with exact evidence2. route the failing slice to the smallest specific skill3. resume this four-app lane only after the blocking slice is repaired

Frequently asked questions