Completion reconciliation — 8 November 2026
Engine outcome delivered: a public local contributor route to launch Neo, find existing work or ask in Q&A, run tests and author a PR without private Neo services.
The operator clarified the repository boundary on 8 November: Agent OS does not belong in the Engine repository. Engine support includes the Neural Link for npx neo-app; GitHub Workflow, Knowledge Base and Memory Core MCP-server setup belong to the Brain or Fleet Manager / Institution line. The June pre-split Agent OS, MCP, seed-path and config-hook prescriptions below are outside this Engine epic's completion scope. They are not claimed repaired here, and no replacement Engine ticket is required.
Delivered work
| Engine acceptance |
Delivered child / evidence |
| Public contributor door and local install/build/test |
#18985, including merged #19337 and the clean-clone receipt: install, browser dependencies, themes, 4,838 tests passed / 2 skipped, clean tree, without Neo/MCP environment. The expired event requirements were explicitly retired as not planned. |
| Current work selection and a usable question route |
#20556, delivered by merged #20557: follow ROADMAP; use existing Q&A when starter work is held or unsuitable; preserve claim policy and public label routes. |
| Public contributor testing and PR authoring without private maintainer initialization |
Skills #310, delivered by merged #311 and published as 0.1.97: declared contributor skill dependencies, materialization and private-tool-free author workflows. Package/CLI and source evidence remain in the parent receipt. |
| Observable launch and outside contribution |
The parent receipt records a separately witnessed styled Grid launch. October measurements record 30 external PRs / 27 merged, 21 authors / 20 first-time. These observations do not attribute causality or claim adoption of a particular Skills version. |
All three native children are closed: two completed, one event close with its delivered contributor door retained and expired requirements explicitly retired. The on-demand contributor route retains the 7 November author decision: no tracked contributor variant, default flip or enrollment selector. Existing maintainer-home prerequisites are not changed by this close.
The later version-specific fresh-adoption watch is not an Engine completion gate. This close makes no fresh-harness, running-Agent-OS, ≤30-minute journey, new conversion, or generated-app cold-run claim. It resolves the delivered Engine contributor workflow.
Historical June framing and pre-split prescriptions — superseded by the completion reconciliation above
Local-first developer onboarding — fork → install → try a lane → PR
Problem Scope
Empirical funnel (V-B-A'd 2026-06-27): 3,217 stars · 214 forks · zero external-authored issues in the recent (Agent-OS-era) tracker. Not an awareness problem and not a no-attempts problem — silent bounce: people fork to try and leave without a word, because there is no newcomer-facing door and the local setup is broken/contradictory. A self-evolving organism that "turns outward" in v13 is only installable if a single dev can fork it and try one lane.
The realistic, honest target (no cloud required, no swarm-A2A promised): a solo dev forks, runs locally on their own inference, and takes one lane end-to-end — pick a good-first-issue → build → open a PR we review on GitHub. The hosted swarm (shared MC, A2A-with-the-team, per-tenant isolation) is a separate private track and stays out of the local onboarding surface.
Intended Solution (facets — sub-tickets open incrementally, NOT hardcoded here)
- Funnel diagnosis (root framing) — measured silent-bounce, the baseline every facet justifies against.
- The front door / "is there a door for me" — a newcomer-facing surface; highest ROI (where the 214 are lost).
- The "ask-here" door — wire the existing unused primitives: the empty
good first issue label, the un-surfaced Q&A Discussions category, an "I'm trying it / how do I…" issue-template path, a CONTRIBUTING that leads with try-a-lane not star-us. Cheapest converter of a silent bounce → a question.
- Setup mechanics (the 3 V-B-A'd config bugs) — Claude Code MCP config triple-contradiction + missing repo-root
.mcp.json; broken seed-script path in 5 docs (ai/scripts/seedAgentIdentities.mjs → ai/scripts/setup/...); github-workflow hardcodes owner:neomjs/repo:neo with no fork override.
- Graceful solo mode — a fresh clone inherits the swarm substrate (the enforcing
NEO_LANE_STATE_ENFORCE=1 stop-hook + swarm-centric AGENTS_STARTUP + mandatory A2A). Gate via an AiConfig leaf (e.g. useAgentHooks, mirroring aiConfig.mailbox.defaultReplyPolicy) so solo is the default, swarm opts in. The clone/fork contributor path receives the contributor audience of AGENTS.md from the existing Skills generator (generate({audience, repos})), while an enrolled maintainer seat keeps its maintainer projection (Euclid's intake, 6688649831).
Out of Scope
- Hosted / shared-swarm work or swarm-A2A-for-external-devs (a separate private track; never on the public onboarding surface).
- Client specifics (private repos only).
Ownership / Delegation
Filed for the team to parallelize — facets are independently deliverable and should be claimed by peers (@neo-opus-ada / @neo-opus-vega / @neo-gpt). This is DX, not v13.1 self-healing; it must not consume the self-healing focus.
Authored by Grace (Claude Opus 4.8). Origin session 090a68e6-1a28-4b20-a5fd-842ebac3e729.
Completion reconciliation — 8 November 2026
Engine outcome delivered: a public local contributor route to launch Neo, find existing work or ask in Q&A, run tests and author a PR without private Neo services.
The operator clarified the repository boundary on 8 November: Agent OS does not belong in the Engine repository. Engine support includes the Neural Link for
npx neo-app; GitHub Workflow, Knowledge Base and Memory Core MCP-server setup belong to the Brain or Fleet Manager / Institution line. The June pre-split Agent OS, MCP, seed-path and config-hook prescriptions below are outside this Engine epic's completion scope. They are not claimed repaired here, and no replacement Engine ticket is required.Delivered work
All three native children are closed: two completed, one event close with its delivered contributor door retained and expired requirements explicitly retired. The on-demand contributor route retains the 7 November author decision: no tracked contributor variant, default flip or enrollment selector. Existing maintainer-home prerequisites are not changed by this close.
The later version-specific fresh-adoption watch is not an Engine completion gate. This close makes no fresh-harness, running-Agent-OS, ≤30-minute journey, new conversion, or generated-app cold-run claim. It resolves the delivered Engine contributor workflow.
Historical June framing and pre-split prescriptions — superseded by the completion reconciliation above
Local-first developer onboarding — fork → install → try a lane → PR
Problem Scope
Empirical funnel (V-B-A'd 2026-06-27): 3,217 stars · 214 forks · zero external-authored issues in the recent (Agent-OS-era) tracker. Not an awareness problem and not a no-attempts problem — silent bounce: people fork to try and leave without a word, because there is no newcomer-facing door and the local setup is broken/contradictory. A self-evolving organism that "turns outward" in v13 is only installable if a single dev can fork it and try one lane.
The realistic, honest target (no cloud required, no swarm-A2A promised): a solo dev forks, runs locally on their own inference, and takes one lane end-to-end — pick a good-first-issue → build → open a PR we review on GitHub. The hosted swarm (shared MC, A2A-with-the-team, per-tenant isolation) is a separate private track and stays out of the local onboarding surface.
Intended Solution (facets — sub-tickets open incrementally, NOT hardcoded here)
good first issuelabel, the un-surfacedQ&ADiscussions category, an "I'm trying it / how do I…" issue-template path, a CONTRIBUTING that leads with try-a-lane not star-us. Cheapest converter of a silent bounce → a question..mcp.json; broken seed-script path in 5 docs (ai/scripts/seedAgentIdentities.mjs→ai/scripts/setup/...);github-workflowhardcodesowner:neomjs/repo:neowith no fork override.NEO_LANE_STATE_ENFORCE=1stop-hook + swarm-centricAGENTS_STARTUP+ mandatory A2A). Gate via an AiConfig leaf (e.g.useAgentHooks, mirroringaiConfig.mailbox.defaultReplyPolicy) so solo is the default, swarm opts in. The clone/fork contributor path receives the contributor audience ofAGENTS.mdfrom the existing Skills generator (generate({audience, repos})), while an enrolled maintainer seat keeps its maintainer projection (Euclid's intake, 6688649831).Out of Scope
Ownership / Delegation
Filed for the team to parallelize — facets are independently deliverable and should be claimed by peers (@neo-opus-ada / @neo-opus-vega / @neo-gpt). This is DX, not v13.1 self-healing; it must not consume the self-healing focus.
Authored by Grace (Claude Opus 4.8). Origin session 090a68e6-1a28-4b20-a5fd-842ebac3e729.