feat(agents): add native plugins for Codex Claude and Cursor

[vm]

Summary

• Add native Hack plugins for Codex, Claude Code, and Cursor.
• Gate all automatic, setup-sync, and interactive onboarding cleanup on the corresponding plugin being installed and enabled.
• Preserve project/user legacy rules, hooks, skills, and MCP entries while a client, plugin, or enabled state is missing.
• Remove only exact generated copies after readiness; preserve customized artifacts.
• Report unavailable plugins as warnings with non-success exit status instead of successful installs.
• Use Cursor's documented /add-plugin flow and Settings > Plugins guidance.
• Share plugin inspection, preparation, cleanup aggregation, and setup command orchestration while retaining client-specific parsing and discovery.

Verification

• 967 tests passed, 5 skipped, 0 failed; 3,275 assertions.
• Added a 36-scenario cutover matrix: missing / disabled / enabled by project / user by generated / customized, across Codex, Claude Code, and Cursor, including MCP preservation.
• Added direct automatic-sync gating and setup-result tests.
• Live missing-client/plugin setup and check commands warn and exit 1; no cleanup runs.
• Current stable Codex CLI successfully added the local marketplace, installed and enabled hack@hack-dance, and reported the expected source. Test registry entries were removed afterward.
• Current stable Claude Code successfully added the local marketplace, installed and enabled hack@hack-dance, exposed hack mcp serve, and installed both skills, hooks, and the rule. Test registry entries were removed afterward.
• Cursor's official 2.5 plugin documentation confirms /add-plugin. Cursor Agent CLI is not installed on the validation host, so a live Cursor CLI install remains a reviewer/manual verification item.
• Ultracite, CLI and DB TypeScript checks, CLI build, generated CLI reference, generated plugin drift check, Codex plugin validation, Claude plugin validation, and git diff checks passed.

Release Signal

• Commit / squash title: feat(agents): add native plugins for Codex Claude and Cursor
• Release intent: feat
• Should this PR trigger a release signal? yes
• No changeset or equivalent release-artifact mechanism is configured in this repository; the Conventional Commit PR title carries the feature release signal.

Semantic Surfaces

• Does this change affect hack run, hack exec, env resolution, runtime-state reconciliation, or lifecycle shell/process behavior? no

Risks / Follow-up

• Plugin list JSON differs between clients; each parser remains client-owned and has targeted state tests.
• Legacy cleanup is readiness-gated and only removes exact generated content or registered historical fingerprints.
• A live Cursor install still needs validation on a host with current Cursor and Cursor Agent CLI; the documented editor flow is verified against Cursor's official current documentation.

[roodboi]

Requesting changes because the migration currently has no safe cutover boundary: a normal interactive Hack command can remove the working standalone integration before the native plugin is installed or enabled.

What I verified on this head:

• full bun test exits successfully
• the focused plugin/setup suite passes (48 tests, 193 assertions)
• bun run typecheck, bun run check, and git diff --check pass
• hosted CI is green across test, runtime-images, Docker E2E, and secret scan
• live codex plugin list --json and claude plugin list --json output match the new parsers

What still needs direct verification before merge:

1. Fresh marketplace add + plugin install for Codex, Claude Code, and Cursor using current stable clients.
2. A cutover matrix for plugin missing / disabled / enabled × project / user legacy artifacts × generated / customized artifacts. Missing or disabled must retain the working legacy integration; enabled may remove only exact generated copies.
3. After a new session, prove both skills, the Claude hooks / Cursor rule, and hack mcp serve are actually loaded from each installed plugin.
4. Exercise interactive hack init and hack setup output so missing/disabled plugins are warnings with non-success status.

Please also complete the PR description's Summary, Verification, Release Signal, Semantic Surfaces, and Risks sections. This is a user-facing feat and the release decision must be explicit per repository policy.

[roodboi]

[P2] Validate and document the current Cursor installation surface. Cursor's current official plugin documentation says to install in the editor with /add-plugin (or Settings → Plugins), while /plugin is not documented: https://cursor.com/changelog/2-5. Because this is the only recovery path printed after cleanup, an invalid slash command strands the user without either integration. Update the canonical guidance and all generated copies, then verify on current stable Cursor and Cursor Agent CLI that the marketplace is accepted, Hack installs/enables, and its rules, skills, and MCP server appear after a new session.

[roodboi]

Re-reviewed the two new commits (e21feee and 6808a21). The original unsafe-cleanup, missing-plugin success, Cursor guidance, and PR-description findings are substantially addressed.

What I verified on this head:

• focused plugin/cutover/setup suite: 36 tests, 217 assertions, 0 failures
• direct CLI TypeScript check and direct Ultracite check on all changed TS/test files
• CLI build and git diff --check
• the readiness matrix preserves legacy content while plugins are missing/disabled

Two false-green cases remain when the plugin is enabled but customized legacy content is preserved; see the inline comments. Please test enabled-plugin cutover with a customized primary artifact and customized MCP entry for all three clients, both scopes, through hack setup <client>, hack setup sync --all-scopes, automatic sync, and interactive hack init. Every path must preserve the customization, warn, and exit nonzero until the duplicate legacy integration is manually reconciled.

Hosted CI has not executed: run 243 is action_required with zero jobs, consistent with the fork workflow awaiting maintainer approval. Please approve/run CI and get it green. A live current-stable Cursor + Cursor Agent CLI marketplace/install/load check also remains outstanding before approval.

[roodboi]

CI is now running, and the main test/typecheck/build, runtime-image, and secret-scan jobs pass. Docker E2E fails deterministically in agent-docs-sync because this PR changed missing native plugins to a non-success sync result without updating the existing E2E contract. See the inline comment for the required coverage. After updating it, run bun run test:e2e:local:docker and keep the earlier customized-artifact cutover cases in scope.

[roodboi]

Re-reviewed 7f72688 plus the merge head 3d71e07.

Verified in an isolated checkout:

• focused plugin/cutover/setup suite: 27 tests, 307 assertions, 0 failures
• agent-docs-sync E2E passes locally, and hosted run 247's docker-e2e job is green
• CLI TypeScript, changed-file Ultracite, CLI build, and git diff --check pass
• customized primary and MCP artifacts are preserved and return non-success across all three clients and both scopes

The prior functional findings are fixed. One merge-generated artifact drift remains: the 3.5.2 merge left the new plugin manifests and bundled Codex skill at 3.5.1, producing four deterministic test failures. See the inline comment.

After regenerating, please run bun test tests/claude-plugin.test.ts tests/cursor-plugin.test.ts tests/codex-plugin.test.ts tests/agent-instruction-source.test.ts and get the hosted test job green. The existing live current-stable Cursor install/load verification thread also remains open.

[roodboi]

[P1] Regenerate the native-plugin artifacts after merging the 3.5.2 release. package.json is now 3.5.2, but this manifest and the Cursor/Codex manifests still declare 3.5.1, and plugins/hack/skills/hack-cli/SKILL.md still embeds v3.5.1 guidance. On 3d71e07, bun test tests/claude-plugin.test.ts tests/cursor-plugin.test.ts tests/codex-plugin.test.ts tests/agent-instruction-source.test.ts fails exactly those four parity checks (the same root cause already failed run 247's test job). Run bun run generate:agent-plugins, commit every generated artifact, rerun that focused command, and get the hosted test job green.