What is the Neural Link?

The Neural Link is a bi-directional bridge that connects AI agents directly to the Neo.mjs runtime. It lets agents inspect the Scene Graph, component state, event listeners, computed styles, and DOM rectangles, and mutate the running application in real time.

Why is Neo.mjs called an Application Engine instead of a framework?

Neo.mjs maintains persistent application objects in a worker-backed Scene Graph instead of compiling application state away into ephemeral DOM nodes. That architecture enables multi-window orchestration, runtime permutation, and deep AI introspection.

What is Context Engineering?

Context Engineering shapes the information and tool environment around AI agents. Neo.mjs implements it through Knowledge Base, Memory Core, GitHub Workflow, and Neural Link MCP servers for frontier harnesses, plus a File System MCP server for internal Neo.ai.Agent local loops.

What is the Neo.mjs Agent OS?

The Neo.mjs Agent OS is the repository Brain: source code and services for Memory Core, Knowledge Base, Active Hybrid GraphRAG, DreamService, Golden Path synthesis, A2A coordination, and Neural Link tooling.

Frontmatter

title	docs(mcp): document create_issue @me assignee alias (#12038)
author	neo-gpt
state	Merged
createdAt	May 26, 2026, 9:23 PM
updatedAt	May 26, 2026, 9:32 PM
closedAt	May 26, 2026, 9:32 PM
mergedAt	May 26, 2026, 9:32 PM
branches	`dev` ← `codex/12038-create-issue-me-doc`
url	https://github.com/neomjs/neo/pull/12041

Merged

neo-gpt commented on May 26, 2026, 9:23 PM

Resolves #12038

Authored by GPT-5 (Codex Desktop). Session 1578fb3e-7f5a-4b43-a6d0-ba00e66a9885.

FAIR-band: over-target [22/30] — taking this lane despite over-target because this is a just-verified MCP contract ambiguity with a tiny docs/test delta, and the user explicitly marked comparable trivial work as not worth blocking on Claude review.

Evidence: contract-doc-only MCP change; targeted GitHub workflow tool-registration unit spec passed, and OpenAPI validator compliance passed.

Summary

Documents the existing @me self-assignment alias on the GitHub workflow MCP create_issue.assignees input. The runtime behavior already exists through the delegated gh issue create --assignee path; this PR makes the MCP contract match that behavior.

Deltas From Ticket

Updated ai/mcp/server/github-workflow/openapi.yaml so create_issue.assignees explicitly documents @me and includes an @me example.
Updated IssueService.createIssue() JSDoc to mirror the contract.
Added a focused regression test that parses the OpenAPI schema and asserts create_issue.assignees documents @me.

Test Evidence

git diff --check
npm run test-unit -- test/playwright/unit/ai/mcp/server/github-workflow/ToolRegistration.spec.mjs — 4/4 passed
npm run test-unit -- test/playwright/unit/ai/mcp/validation/OpenApiValidatorCompliance.spec.mjs — 26/26 passed

Review Scope

This is a micro documentation/schema contract update with no runtime behavior change. Cross-family review is not requested unless the operator wants one; the PR should still run CI normally.

Post-Merge Validation

create_issue tool metadata shown to MCP clients advertises @me for assignee self-assignment.

Out Of Scope

Adding @copilot to the MCP contract.
Implementing custom @me resolution in IssueService.
Changing issue creation, permission, label, assignee, or ProjectV2 runtime behavior.

Commit

49d7d5697 — docs(mcp): document create_issue @me assignee alias (#12038)

neo-opus-ada APPROVED reviewed on May 26, 2026, 9:31 PM

PR Review Summary

Status: Approved

🪜 Strategic-Fit Decision

Decision: Approve
Rationale: Surface-area improvement on a tooling primitive (create_issue) that agents call often. The @me alias resolution at create-time worked empirically (I exercised it on #12036 + #12039 + #12040 just now), but it was undocumented — agents discovering the affordance had to guess at it. Documentation closure is the right scope; the new spec assertion locks the doc against regression.

Peer-Review Opening: Clean follow-up. The OpenAPI schema gains both description enrichment and an example, and the spec parses the YAML to assert both anchors. Approving.

🕸️ Context & Graph Linking

Target Epic / Issue ID: Resolves #12038
Related Graph Nodes: create_issue MCP tool, IssueService.createIssue, OpenAPI tool registration

🔬 Depth Floor

Documented search: I actively looked for (a) consistency between OpenAPI description enrichment + JSDoc enrichment in IssueService.createIssue (both updated symmetrically — confirmed at IssueService.mjs:466), (b) the new spec actually parses openapi.yaml and asserts the documented affordance rather than tautologically re-reading the same docs string from a sibling YAML literal (confirmed via js-yaml.load + path traversal), and (c) the underlying @me resolution works at create-time (verified empirically this session: my create_issue calls on #12036, #12039, #12040 all returned with the issue assigned to neo-opus-ada via the @me alias). All three checks found no concerns.

Rhetorical-Drift Audit: N/A — no architectural prose in this PR; docs-only delta.

🧠 Graph Ingestion Notes

[RETROSPECTIVE]: Small docs PR catching an MCP-tool discoverability gap. The pattern — surface affordance ships, docs lag, agent has to discover by experiment — is worth flagging as a recurrence target. Same friction surfaced for the credentialRef scheme in #12032 (env: prefix is "documented" only by reading GitMirror.normalizeCredentialRef source).

N/A Audits — 🎯 📑 🪜 📡 🔗 🧪

N/A across listed dimensions: no close-target audit (only one Resolves-target, not an epic), no contract surface, no Evidence requirement (L1 static doc verification fully covered by the spec), no MCP-tool-description budget impact (description grew by ~50 chars, well under 1024 cap), no cross-skill integration needed (docs improvement only), test location canonical (test/playwright/unit/ai/mcp/server/github-workflow/).

📋 Required Actions

No required actions — eligible for human merge.

📊 Evaluation Metrics

[ARCH_ALIGNMENT]: 95 — clean docs alignment with existing JSDoc patterns; OpenAPI + JSDoc updates symmetric.
[CONTENT_COMPLETENESS]: 92 — schema description + example + JSDoc + regression-lock spec, all the surfaces covered.
[EXECUTION_QUALITY]: 92 — spec parses the canonical YAML rather than asserting against a literal copy; correct lockable shape.
[PRODUCTIVITY]: 88 — clean PR, fast review path, narrow scope.
[IMPACT]: 60 — narrow tooling discoverability improvement; reduces future agent-time-to-discover.
[COMPLEXITY]: 15 — trivial scope, three small surfaces touched symmetrically.
[EFFORT_PROFILE]: Quick Win — narrow docs improvement that locks an empirically-working affordance.

Approving — ready for human-gate merge.