[PR #1549] [MERGED] test(clients): verify native Claude Code integration #2964

New issue

Closed

opened 2026-10-17 21:12:18 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-17 21:12:18 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/morluto/rea/pull/1549
Author: @N0zoM1z0
Created: 10/10/2026
Status: ✅ Merged
Merged: 10/10/2026
Merged by: @N0zoM1z0

Base: main ← Head: test/claude-native-client

📝 Commits (1)

e657637 test(clients): verify native Claude Code integration

📊 Changes

4 files changed (+541 additions, -0 deletions)

View changed files

📝 docs/testing.md (+14 -0)
📝 package.json (+1 -0)
➕ scripts/verify-claude-client.mjs (+368 -0)
➕ scripts/verify/clients/anthropic-model-fixture.mjs (+158 -0)

📄 Description

REA configures Claude Code, but native skill loading and analysis round trips lacked a reproducible verification lane. Add verify:claude-client with the actual installed client and a loopback Anthropic Messages/SSE fixture.

Exercise automatic user-profile MCP discovery, native WaitForMcpServers during asynchronous startup, the complete installed skill body through Skill, all advertised REA input schemas and full named-schema JavaScript Evidence for a Unicode path. Check public setup plans, preserved caller preferences, backups, idempotence, native final replies and owned-process cleanup. Keep native request bodies and resource-hint envelopes as artifacts; do not retain authentication headers.

Validation: actual @anthropic-ai/claude-code@2.1.296 chat/call on source and a fresh private production-only installed tarball; 139 tools and 139 forwarded input schemas per workflow; npm run check:fast. Linux x64, isolated CLAUDE_CONFIG_DIR, independent Git project, user settings and default built-in tools, synthetic local-model usage. Live Anthropic API, native Windows, other settings scopes and bare-mode personal-skill activation are outside this lane. Product runtime and contracts are unchanged; no fresh Hopper/Ghidra verification.

Producer references: pinned Claude Code release, official Messages/SSE types. CLI flags, skill and pending-server behavior were checked against the installed native client.

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/morluto/rea/pull/1549 **Author:** [@N0zoM1z0](https://github.com/N0zoM1z0) **Created:** 10/10/2026 **Status:** ✅ Merged **Merged:** 10/10/2026 **Merged by:** [@N0zoM1z0](https://github.com/N0zoM1z0) **Base:** `main` ← **Head:** `test/claude-native-client` --- ### 📝 Commits (1) - [`e657637`](https://github.com/morluto/rea/commit/e657637f0d29b58ded675f037e7d8b4215afbcfa) test(clients): verify native Claude Code integration ### 📊 Changes **4 files changed** (+541 additions, -0 deletions) <details> <summary>View changed files</summary> 📝 `docs/testing.md` (+14 -0) 📝 `package.json` (+1 -0) ➕ `scripts/verify-claude-client.mjs` (+368 -0) ➕ `scripts/verify/clients/anthropic-model-fixture.mjs` (+158 -0) </details> ### 📄 Description REA configures Claude Code, but native skill loading and analysis round trips lacked a reproducible verification lane. Add `verify:claude-client` with the actual installed client and a loopback Anthropic Messages/SSE fixture. Exercise automatic user-profile MCP discovery, native `WaitForMcpServers` during asynchronous startup, the complete installed skill body through `Skill`, all advertised REA input schemas and full named-schema JavaScript Evidence for a Unicode path. Check public setup plans, preserved caller preferences, backups, idempotence, native final replies and owned-process cleanup. Keep native request bodies and resource-hint envelopes as artifacts; do not retain authentication headers. Validation: actual `@anthropic-ai/claude-code@2.1.296` chat/call on source and a fresh private production-only installed tarball; 139 tools and 139 forwarded input schemas per workflow; `npm run check:fast`. Linux x64, isolated `CLAUDE_CONFIG_DIR`, independent Git project, user settings and default built-in tools, synthetic local-model usage. Live Anthropic API, native Windows, other settings scopes and bare-mode personal-skill activation are outside this lane. Product runtime and contracts are unchanged; no fresh Hopper/Ghidra verification. Producer references: [pinned Claude Code release](https://github.com/anthropics/claude-code/tree/2301018b1f61073c501a8e7a4813ef48c239163b), [official Messages/SSE types](https://github.com/anthropics/anthropic-sdk-typescript/blob/2979ed5cef4b2f059b95af23e08728c01e588f86/src/resources/messages/messages.ts). CLI flags, skill and pending-server behavior were checked against the installed native client. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-10-17 21:12:18 +00:00

closed this issue
added the
pull-request
label

bsakaguchi referenced this issue

2026-10-31 01:07:18 +00:00

[PR #2964] [MERGED] test(reconstruction): declare complete comparison fixture facets #5521

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

morluto/rea#2964

No description provided.

Rows
Columns