fix: unblock Archon when launched from a Claude Code terminal and on Bun 1.3+#1097
Closed
samuelcombey wants to merge 1 commit intocoleam00:devfrom
Closed
fix: unblock Archon when launched from a Claude Code terminal and on Bun 1.3+#1097samuelcombey wants to merge 1 commit intocoleam00:devfrom
samuelcombey wants to merge 1 commit intocoleam00:devfrom
Conversation
…Bun 1.3+ Two blockers hit during a fresh setup from inside a Claude Code session on Bun 1.3.12; both reproduce deterministically. 1) Claude Agent SDK subprocess refuses to launch (nested-session guard). The parent Claude Code shell exports CLAUDECODE, CLAUDE_CODE_ENTRYPOINT, CLAUDE_CODE_EXECPATH, CLAUDE_CODE_HIDE_ACCOUNT_INFO, CLAUDE_CODE_NO_FLICKER, and CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS. Archon's existing cleanup in buildSubprocessEnv() only stripped CLAUDECODE, and the SDK leaks process.env into the spawned child regardless of the explicit `env` option passed to query() — so the subprocess sees the remaining markers, hits its own nested-session guard, and the orchestrator hangs after `using_global_auth` with no error. Strip all six markers from process.env at server and CLI entry points so nothing can leak. Auth vars (OAUTH_TOKEN, USE_BEDROCK, USE_VERTEX) are kept. 2) Telegram adapter dies on first polling error on Bun 1.3+. Telegraf's redactToken() does `error.message = error.message.replace(...)` to sanitize bot tokens from error stacks. On Bun 1.3+, Error.message on fetch/network error instances is non-writable, so the assignment throws TypeError, kills the polling loop, and leaves the adapter silently dead with `telegram.start_failed`. Patch telegraf@4.16.3 via `bun patch` to wrap the assignment in try/catch; the original error is still re-thrown so Telegraf's own retry/reconnect logic runs normally. Also picks up an incidental bun.lock re-sync from `bun patch --commit`: the workspace package versions in the lockfile were stale (0.1.0/0.2.0) while the actual package.json files have been 0.3.5 for a while.
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: ⛔ Files ignored due to path filters (1)
📒 Files selected for processing (4)
📝 WalkthroughWalkthroughThis PR adds environment variable cleanup at startup to prevent nested-session context detection in spawned subprocesses, removes Claude Code markers (CLAUDECODE and related variables) from process.env in CLI and server entry points, and patches telegraf@4.16.3 to safely handle non-writable error messages in Bun environments. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~12 minutes Possibly related issues
Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: f41f4e8917
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| + // original error rather than killing the polling loop. | ||
| + try { | ||
| + error.message = error.message.replace(/\/(bot|user)(\d+):[^/]+\//, '/$1$2:[REDACTED]/'); | ||
| + } catch (_) { /* message is read-only; re-throw original */ } |
There was a problem hiding this comment.
Catching and ignoring failures in redactToken means Bun readonly-message errors are rethrown with the original unredacted text, which can include /bot<id>:<token>/ and leak the Telegram bot token into logs. In this exact Bun 1.3+ path, the new behavior fixes retries but drops the security guarantee this function exists to provide; please keep retry behavior while still throwing a redacted error payload.
Useful? React with 👍 / 👎.
Wirasm
added a commit
that referenced
this pull request
Apr 12, 2026
…marker strip, tests 1. Align dotenv to ^17 (was ^16, rest of monorepo uses ^17.2.3) 2. Remove incorrect SUBPROCESS_ENV_ALLOWLIST claim from docs — the SDK bypasses the env option and uses process.env directly (#1097) 3. Add CLAUDECODE=1 warning to server entry point (was only in CLI) 4. Add diagnostic payload content test for withFirstMessageTimeout 5. Integrate #1097's finding: strip CLAUDECODE + CLAUDE_CODE_* session markers (except auth vars) + NODE_OPTIONS + VSCODE_INSPECTOR_OPTIONS from process.env at entry point. Pattern-matched on CLAUDE_CODE_* prefix rather than hardcoding 6 names, so future Claude Code markers are handled automatically. Auth vars (CLAUDE_CODE_OAUTH_TOKEN, CLAUDE_CODE_USE_BEDROCK, CLAUDE_CODE_USE_VERTEX) are preserved. Root cause per #1097: the Claude Agent SDK leaks process.env into the spawned child regardless of the explicit env option, so the only way to prevent the nested-session deadlock is to delete the markers from process.env at the entry point. Validation: bun run validate passes, 125 paths tests (6 new marker tests), 60 claude tests (1 new diagnostic test), DATABASE_URL leak verified stripped (target repo .env DATABASE_URL does not affect Archon DB selection).
Wirasm
added a commit
that referenced
this pull request
Apr 12, 2026
…only CWD The allowlist was wrong for a single-developer tool: - It blocked keys the user intentionally set in ~/.archon/.env (ANTHROPIC_API_KEY, AWS_*, CLAUDE_CONFIG_DIR, MiniMax vars, etc.) - It was bypassed by the SDK anyway (process.env leaks to subprocess regardless of the env option — see #1097) - It attracted a constant stream of PRs adding keys (#1060, #1093, #1099) New model: CWD .env keys are the only untrusted source. stripCwdEnv() at entry point handles that. Everything in ~/.archon/.env + shell env passes through to the subprocess. No filtering, no second-guessing. Changes: - Delete env-allowlist.ts and env-allowlist.test.ts - Simplify buildSubprocessEnv() to return { ...process.env } with auth-mode logging (no token stripping — user controls their config) - Replace 4 allowlist-based tests with 1 pass-through test - Remove env-allowlist.test.ts from core test batch - Update security.md and cli.md docs to reflect the new model The CLAUDECODE + CLAUDE_CODE_* marker strip and NODE_OPTIONS strip remain in stripCwdEnv() at entry point — those are process-level safety (not per-subprocess filtering) and are needed regardless.
Collaborator
|
Thanks for the excellent debugging on the nested Claude Code session issue — your finding that the SDK leaks We've integrated your The Telegraf Bun 1.3+ fix in this PR is still needed — #1092 doesn't address that. Would you be willing to resubmit a PR with just the Telegraf Summary:
|
Wirasm
added a commit
that referenced
this pull request
Apr 12, 2026
…t timeout (#1067, #1030, #1098, #1070) * fix: strip CWD .env leak, enable platform adapters in serve, add first-event timeout (#1067) Three bugs fixed: (1) Bun auto-loads CWD .env files before user code, leaking non-overlapping keys into the Archon process — new stripCwdEnv() boot import removes them before any module reads env. (2) archon serve hardcoded skipPlatformAdapters:true, preventing Slack/Telegram/Discord from starting. (3) Claude SDK query had no first-event timeout, causing silent 30-min hangs when the subprocess wedges — new withFirstMessageTimeout wrapper races the first event against a configurable deadline (default 60s). Changes: - Add @archon/paths/strip-cwd-env and strip-cwd-env-boot modules - Import boot module as first import in CLI entry point - Remove skipPlatformAdapters: true from serve.ts - Add withFirstMessageTimeout + diagnostics to ClaudeClient - Add CLAUDECODE=1 nested-session warning to CLI - Add 9 unit tests (6 strip-cwd-env + 3 timeout) Fixes #1067 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: address review findings for PR #1092 Fixed: - Clear setTimeout timer in withFirstMessageTimeout finally block (HIGH-1) - Add strip-cwd-env-boot to server/src/index.ts for direct dev:server path (MEDIUM-1) - Warn to stderr on non-ENOENT errors in stripCwdEnv (MEDIUM-2) - Update stale configuration.md docs for new env-loading mechanism (HIGH-2) - Add ARCHON_CLAUDE_FIRST_EVENT_TIMEOUT_MS and ARCHON_SUPPRESS_NESTED_CLAUDE_WARNING env vars to docs - Add nested Claude Code hang troubleshooting entry - Fix boot module JSDoc: "CLI and server" → "CLI" only - Fix stripCwdEnv JSDoc: remove stale "override: true" reference - Update .claude/rules/cli.md startup behavior section - Update CLAUDE.md @archon/paths description with new exports Tests added: - Assert controller.signal.aborted on timeout - Handle generator that completes immediately without yielding - Strip distinct keys from different .env files Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * simplify: replace string sentinel with typed error class in withFirstMessageTimeout Replace the '__timeout__' string sentinel used to identify timeout rejections with a dedicated FirstEventTimeoutError class. instanceof checks are more explicit and robust than string comparison on error messages. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: address review findings — dotenv version, docs, server warning, marker strip, tests 1. Align dotenv to ^17 (was ^16, rest of monorepo uses ^17.2.3) 2. Remove incorrect SUBPROCESS_ENV_ALLOWLIST claim from docs — the SDK bypasses the env option and uses process.env directly (#1097) 3. Add CLAUDECODE=1 warning to server entry point (was only in CLI) 4. Add diagnostic payload content test for withFirstMessageTimeout 5. Integrate #1097's finding: strip CLAUDECODE + CLAUDE_CODE_* session markers (except auth vars) + NODE_OPTIONS + VSCODE_INSPECTOR_OPTIONS from process.env at entry point. Pattern-matched on CLAUDE_CODE_* prefix rather than hardcoding 6 names, so future Claude Code markers are handled automatically. Auth vars (CLAUDE_CODE_OAUTH_TOKEN, CLAUDE_CODE_USE_BEDROCK, CLAUDE_CODE_USE_VERTEX) are preserved. Root cause per #1097: the Claude Agent SDK leaks process.env into the spawned child regardless of the explicit env option, so the only way to prevent the nested-session deadlock is to delete the markers from process.env at the entry point. Validation: bun run validate passes, 125 paths tests (6 new marker tests), 60 claude tests (1 new diagnostic test), DATABASE_URL leak verified stripped (target repo .env DATABASE_URL does not affect Archon DB selection). * refactor: remove SUBPROCESS_ENV_ALLOWLIST — trust user config, strip only CWD The allowlist was wrong for a single-developer tool: - It blocked keys the user intentionally set in ~/.archon/.env (ANTHROPIC_API_KEY, AWS_*, CLAUDE_CONFIG_DIR, MiniMax vars, etc.) - It was bypassed by the SDK anyway (process.env leaks to subprocess regardless of the env option — see #1097) - It attracted a constant stream of PRs adding keys (#1060, #1093, #1099) New model: CWD .env keys are the only untrusted source. stripCwdEnv() at entry point handles that. Everything in ~/.archon/.env + shell env passes through to the subprocess. No filtering, no second-guessing. Changes: - Delete env-allowlist.ts and env-allowlist.test.ts - Simplify buildSubprocessEnv() to return { ...process.env } with auth-mode logging (no token stripping — user controls their config) - Replace 4 allowlist-based tests with 1 pass-through test - Remove env-allowlist.test.ts from core test batch - Update security.md and cli.md docs to reflect the new model The CLAUDECODE + CLAUDE_CODE_* marker strip and NODE_OPTIONS strip remain in stripCwdEnv() at entry point — those are process-level safety (not per-subprocess filtering) and are needed regardless. * fix: restore override:true for archon env, add integration tests The integration tests caught a real issue: without override:true, the ~/.archon/.env load doesn't win over shell-inherited env vars. If the user's shell profile exports PORT=9999 and ~/.archon/.env has PORT=3000, the user expects Archon to use 3000. stripCwdEnv() handles CWD .env files (untrusted). override:true handles shell-inherited vars (trusted but less specific than ~/.archon/.env). Different concerns, both needed. Also adds 6 integration tests covering the full entry-point flow: 1. Global auth user with ANTHROPIC_API_KEY in CWD .env — stripped 2. OAuth token in archon env + random key in CWD — CWD stripped, archon kept 3. General leak test — nothing from CWD reaches subprocess 4. Same key in both CWD and archon — archon value wins 5. CLAUDECODE markers stripped even when not from CWD .env 6. CLAUDE_CODE_OAUTH_TOKEN survives marker strip * test: add DATABASE_URL leak scenarios to env integration tests * fix: move CLAUDECODE warning into stripCwdEnv, remove dead useGlobalAuth logic Review findings addressed: 1. CLAUDECODE warning was dead code — the boot import deleted CLAUDECODE from process.env before the warning check in cli.ts/server/index.ts could fire. Moved the warning into stripCwdEnv() itself, emitted BEFORE the deletion. Removed duplicate warning code from both entry points. 2. useGlobalAuth token stripping removed (intentional, not regression) — the old code stripped CLAUDE_CODE_OAUTH_TOKEN and CLAUDE_API_KEY when useGlobalAuth=true. Per design discussion: the user controls ~/.archon/.env and all keys they set are intentional. If they want global auth, they just don't set tokens. Simplified buildSubprocessEnv to log auth mode for diagnostics only, no filtering. 3. Docs "no override needed" corrected — cli.md and configuration.md now reflect the actual code (override: true). --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: Rasmus Widing <rasmus.widing@gmail.com>
Collaborator
|
Closing — this PR bundles two unrelated fixes:
Thanks for identifying both issues — the marker stripping analysis directly informed the structural fix that landed. |
prospapledge88
pushed a commit
to prospapledge88/Archon
that referenced
this pull request
Apr 14, 2026
…t timeout (coleam00#1067, coleam00#1030, coleam00#1098, coleam00#1070) * fix: strip CWD .env leak, enable platform adapters in serve, add first-event timeout (coleam00#1067) Three bugs fixed: (1) Bun auto-loads CWD .env files before user code, leaking non-overlapping keys into the Archon process — new stripCwdEnv() boot import removes them before any module reads env. (2) archon serve hardcoded skipPlatformAdapters:true, preventing Slack/Telegram/Discord from starting. (3) Claude SDK query had no first-event timeout, causing silent 30-min hangs when the subprocess wedges — new withFirstMessageTimeout wrapper races the first event against a configurable deadline (default 60s). Changes: - Add @archon/paths/strip-cwd-env and strip-cwd-env-boot modules - Import boot module as first import in CLI entry point - Remove skipPlatformAdapters: true from serve.ts - Add withFirstMessageTimeout + diagnostics to ClaudeClient - Add CLAUDECODE=1 nested-session warning to CLI - Add 9 unit tests (6 strip-cwd-env + 3 timeout) Fixes coleam00#1067 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: address review findings for PR coleam00#1092 Fixed: - Clear setTimeout timer in withFirstMessageTimeout finally block (HIGH-1) - Add strip-cwd-env-boot to server/src/index.ts for direct dev:server path (MEDIUM-1) - Warn to stderr on non-ENOENT errors in stripCwdEnv (MEDIUM-2) - Update stale configuration.md docs for new env-loading mechanism (HIGH-2) - Add ARCHON_CLAUDE_FIRST_EVENT_TIMEOUT_MS and ARCHON_SUPPRESS_NESTED_CLAUDE_WARNING env vars to docs - Add nested Claude Code hang troubleshooting entry - Fix boot module JSDoc: "CLI and server" → "CLI" only - Fix stripCwdEnv JSDoc: remove stale "override: true" reference - Update .claude/rules/cli.md startup behavior section - Update CLAUDE.md @archon/paths description with new exports Tests added: - Assert controller.signal.aborted on timeout - Handle generator that completes immediately without yielding - Strip distinct keys from different .env files Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * simplify: replace string sentinel with typed error class in withFirstMessageTimeout Replace the '__timeout__' string sentinel used to identify timeout rejections with a dedicated FirstEventTimeoutError class. instanceof checks are more explicit and robust than string comparison on error messages. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: address review findings — dotenv version, docs, server warning, marker strip, tests 1. Align dotenv to ^17 (was ^16, rest of monorepo uses ^17.2.3) 2. Remove incorrect SUBPROCESS_ENV_ALLOWLIST claim from docs — the SDK bypasses the env option and uses process.env directly (coleam00#1097) 3. Add CLAUDECODE=1 warning to server entry point (was only in CLI) 4. Add diagnostic payload content test for withFirstMessageTimeout 5. Integrate coleam00#1097's finding: strip CLAUDECODE + CLAUDE_CODE_* session markers (except auth vars) + NODE_OPTIONS + VSCODE_INSPECTOR_OPTIONS from process.env at entry point. Pattern-matched on CLAUDE_CODE_* prefix rather than hardcoding 6 names, so future Claude Code markers are handled automatically. Auth vars (CLAUDE_CODE_OAUTH_TOKEN, CLAUDE_CODE_USE_BEDROCK, CLAUDE_CODE_USE_VERTEX) are preserved. Root cause per coleam00#1097: the Claude Agent SDK leaks process.env into the spawned child regardless of the explicit env option, so the only way to prevent the nested-session deadlock is to delete the markers from process.env at the entry point. Validation: bun run validate passes, 125 paths tests (6 new marker tests), 60 claude tests (1 new diagnostic test), DATABASE_URL leak verified stripped (target repo .env DATABASE_URL does not affect Archon DB selection). * refactor: remove SUBPROCESS_ENV_ALLOWLIST — trust user config, strip only CWD The allowlist was wrong for a single-developer tool: - It blocked keys the user intentionally set in ~/.archon/.env (ANTHROPIC_API_KEY, AWS_*, CLAUDE_CONFIG_DIR, MiniMax vars, etc.) - It was bypassed by the SDK anyway (process.env leaks to subprocess regardless of the env option — see coleam00#1097) - It attracted a constant stream of PRs adding keys (coleam00#1060, coleam00#1093, coleam00#1099) New model: CWD .env keys are the only untrusted source. stripCwdEnv() at entry point handles that. Everything in ~/.archon/.env + shell env passes through to the subprocess. No filtering, no second-guessing. Changes: - Delete env-allowlist.ts and env-allowlist.test.ts - Simplify buildSubprocessEnv() to return { ...process.env } with auth-mode logging (no token stripping — user controls their config) - Replace 4 allowlist-based tests with 1 pass-through test - Remove env-allowlist.test.ts from core test batch - Update security.md and cli.md docs to reflect the new model The CLAUDECODE + CLAUDE_CODE_* marker strip and NODE_OPTIONS strip remain in stripCwdEnv() at entry point — those are process-level safety (not per-subprocess filtering) and are needed regardless. * fix: restore override:true for archon env, add integration tests The integration tests caught a real issue: without override:true, the ~/.archon/.env load doesn't win over shell-inherited env vars. If the user's shell profile exports PORT=9999 and ~/.archon/.env has PORT=3000, the user expects Archon to use 3000. stripCwdEnv() handles CWD .env files (untrusted). override:true handles shell-inherited vars (trusted but less specific than ~/.archon/.env). Different concerns, both needed. Also adds 6 integration tests covering the full entry-point flow: 1. Global auth user with ANTHROPIC_API_KEY in CWD .env — stripped 2. OAuth token in archon env + random key in CWD — CWD stripped, archon kept 3. General leak test — nothing from CWD reaches subprocess 4. Same key in both CWD and archon — archon value wins 5. CLAUDECODE markers stripped even when not from CWD .env 6. CLAUDE_CODE_OAUTH_TOKEN survives marker strip * test: add DATABASE_URL leak scenarios to env integration tests * fix: move CLAUDECODE warning into stripCwdEnv, remove dead useGlobalAuth logic Review findings addressed: 1. CLAUDECODE warning was dead code — the boot import deleted CLAUDECODE from process.env before the warning check in cli.ts/server/index.ts could fire. Moved the warning into stripCwdEnv() itself, emitted BEFORE the deletion. Removed duplicate warning code from both entry points. 2. useGlobalAuth token stripping removed (intentional, not regression) — the old code stripped CLAUDE_CODE_OAUTH_TOKEN and CLAUDE_API_KEY when useGlobalAuth=true. Per design discussion: the user controls ~/.archon/.env and all keys they set are intentional. If they want global auth, they just don't set tokens. Simplified buildSubprocessEnv to log auth mode for diagnostics only, no filtering. 3. Docs "no override needed" corrected — cli.md and configuration.md now reflect the actual code (override: true). --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: Rasmus Widing <rasmus.widing@gmail.com>
Tyone88
pushed a commit
to Tyone88/Archon
that referenced
this pull request
Apr 16, 2026
…t timeout (coleam00#1067, coleam00#1030, coleam00#1098, coleam00#1070) * fix: strip CWD .env leak, enable platform adapters in serve, add first-event timeout (coleam00#1067) Three bugs fixed: (1) Bun auto-loads CWD .env files before user code, leaking non-overlapping keys into the Archon process — new stripCwdEnv() boot import removes them before any module reads env. (2) archon serve hardcoded skipPlatformAdapters:true, preventing Slack/Telegram/Discord from starting. (3) Claude SDK query had no first-event timeout, causing silent 30-min hangs when the subprocess wedges — new withFirstMessageTimeout wrapper races the first event against a configurable deadline (default 60s). Changes: - Add @archon/paths/strip-cwd-env and strip-cwd-env-boot modules - Import boot module as first import in CLI entry point - Remove skipPlatformAdapters: true from serve.ts - Add withFirstMessageTimeout + diagnostics to ClaudeClient - Add CLAUDECODE=1 nested-session warning to CLI - Add 9 unit tests (6 strip-cwd-env + 3 timeout) Fixes coleam00#1067 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: address review findings for PR coleam00#1092 Fixed: - Clear setTimeout timer in withFirstMessageTimeout finally block (HIGH-1) - Add strip-cwd-env-boot to server/src/index.ts for direct dev:server path (MEDIUM-1) - Warn to stderr on non-ENOENT errors in stripCwdEnv (MEDIUM-2) - Update stale configuration.md docs for new env-loading mechanism (HIGH-2) - Add ARCHON_CLAUDE_FIRST_EVENT_TIMEOUT_MS and ARCHON_SUPPRESS_NESTED_CLAUDE_WARNING env vars to docs - Add nested Claude Code hang troubleshooting entry - Fix boot module JSDoc: "CLI and server" → "CLI" only - Fix stripCwdEnv JSDoc: remove stale "override: true" reference - Update .claude/rules/cli.md startup behavior section - Update CLAUDE.md @archon/paths description with new exports Tests added: - Assert controller.signal.aborted on timeout - Handle generator that completes immediately without yielding - Strip distinct keys from different .env files Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * simplify: replace string sentinel with typed error class in withFirstMessageTimeout Replace the '__timeout__' string sentinel used to identify timeout rejections with a dedicated FirstEventTimeoutError class. instanceof checks are more explicit and robust than string comparison on error messages. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: address review findings — dotenv version, docs, server warning, marker strip, tests 1. Align dotenv to ^17 (was ^16, rest of monorepo uses ^17.2.3) 2. Remove incorrect SUBPROCESS_ENV_ALLOWLIST claim from docs — the SDK bypasses the env option and uses process.env directly (coleam00#1097) 3. Add CLAUDECODE=1 warning to server entry point (was only in CLI) 4. Add diagnostic payload content test for withFirstMessageTimeout 5. Integrate coleam00#1097's finding: strip CLAUDECODE + CLAUDE_CODE_* session markers (except auth vars) + NODE_OPTIONS + VSCODE_INSPECTOR_OPTIONS from process.env at entry point. Pattern-matched on CLAUDE_CODE_* prefix rather than hardcoding 6 names, so future Claude Code markers are handled automatically. Auth vars (CLAUDE_CODE_OAUTH_TOKEN, CLAUDE_CODE_USE_BEDROCK, CLAUDE_CODE_USE_VERTEX) are preserved. Root cause per coleam00#1097: the Claude Agent SDK leaks process.env into the spawned child regardless of the explicit env option, so the only way to prevent the nested-session deadlock is to delete the markers from process.env at the entry point. Validation: bun run validate passes, 125 paths tests (6 new marker tests), 60 claude tests (1 new diagnostic test), DATABASE_URL leak verified stripped (target repo .env DATABASE_URL does not affect Archon DB selection). * refactor: remove SUBPROCESS_ENV_ALLOWLIST — trust user config, strip only CWD The allowlist was wrong for a single-developer tool: - It blocked keys the user intentionally set in ~/.archon/.env (ANTHROPIC_API_KEY, AWS_*, CLAUDE_CONFIG_DIR, MiniMax vars, etc.) - It was bypassed by the SDK anyway (process.env leaks to subprocess regardless of the env option — see coleam00#1097) - It attracted a constant stream of PRs adding keys (coleam00#1060, coleam00#1093, coleam00#1099) New model: CWD .env keys are the only untrusted source. stripCwdEnv() at entry point handles that. Everything in ~/.archon/.env + shell env passes through to the subprocess. No filtering, no second-guessing. Changes: - Delete env-allowlist.ts and env-allowlist.test.ts - Simplify buildSubprocessEnv() to return { ...process.env } with auth-mode logging (no token stripping — user controls their config) - Replace 4 allowlist-based tests with 1 pass-through test - Remove env-allowlist.test.ts from core test batch - Update security.md and cli.md docs to reflect the new model The CLAUDECODE + CLAUDE_CODE_* marker strip and NODE_OPTIONS strip remain in stripCwdEnv() at entry point — those are process-level safety (not per-subprocess filtering) and are needed regardless. * fix: restore override:true for archon env, add integration tests The integration tests caught a real issue: without override:true, the ~/.archon/.env load doesn't win over shell-inherited env vars. If the user's shell profile exports PORT=9999 and ~/.archon/.env has PORT=3000, the user expects Archon to use 3000. stripCwdEnv() handles CWD .env files (untrusted). override:true handles shell-inherited vars (trusted but less specific than ~/.archon/.env). Different concerns, both needed. Also adds 6 integration tests covering the full entry-point flow: 1. Global auth user with ANTHROPIC_API_KEY in CWD .env — stripped 2. OAuth token in archon env + random key in CWD — CWD stripped, archon kept 3. General leak test — nothing from CWD reaches subprocess 4. Same key in both CWD and archon — archon value wins 5. CLAUDECODE markers stripped even when not from CWD .env 6. CLAUDE_CODE_OAUTH_TOKEN survives marker strip * test: add DATABASE_URL leak scenarios to env integration tests * fix: move CLAUDECODE warning into stripCwdEnv, remove dead useGlobalAuth logic Review findings addressed: 1. CLAUDECODE warning was dead code — the boot import deleted CLAUDECODE from process.env before the warning check in cli.ts/server/index.ts could fire. Moved the warning into stripCwdEnv() itself, emitted BEFORE the deletion. Removed duplicate warning code from both entry points. 2. useGlobalAuth token stripping removed (intentional, not regression) — the old code stripped CLAUDE_CODE_OAUTH_TOKEN and CLAUDE_API_KEY when useGlobalAuth=true. Per design discussion: the user controls ~/.archon/.env and all keys they set are intentional. If they want global auth, they just don't set tokens. Simplified buildSubprocessEnv to log auth mode for diagnostics only, no filtering. 3. Docs "no override needed" corrected — cli.md and configuration.md now reflect the actual code (override: true). --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: Rasmus Widing <rasmus.widing@gmail.com>
joaobmonteiro
pushed a commit
to joaobmonteiro/Archon
that referenced
this pull request
Apr 26, 2026
…t timeout (coleam00#1067, coleam00#1030, coleam00#1098, coleam00#1070) * fix: strip CWD .env leak, enable platform adapters in serve, add first-event timeout (coleam00#1067) Three bugs fixed: (1) Bun auto-loads CWD .env files before user code, leaking non-overlapping keys into the Archon process — new stripCwdEnv() boot import removes them before any module reads env. (2) archon serve hardcoded skipPlatformAdapters:true, preventing Slack/Telegram/Discord from starting. (3) Claude SDK query had no first-event timeout, causing silent 30-min hangs when the subprocess wedges — new withFirstMessageTimeout wrapper races the first event against a configurable deadline (default 60s). Changes: - Add @archon/paths/strip-cwd-env and strip-cwd-env-boot modules - Import boot module as first import in CLI entry point - Remove skipPlatformAdapters: true from serve.ts - Add withFirstMessageTimeout + diagnostics to ClaudeClient - Add CLAUDECODE=1 nested-session warning to CLI - Add 9 unit tests (6 strip-cwd-env + 3 timeout) Fixes coleam00#1067 Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com> * fix: address review findings for PR coleam00#1092 Fixed: - Clear setTimeout timer in withFirstMessageTimeout finally block (HIGH-1) - Add strip-cwd-env-boot to server/src/index.ts for direct dev:server path (MEDIUM-1) - Warn to stderr on non-ENOENT errors in stripCwdEnv (MEDIUM-2) - Update stale configuration.md docs for new env-loading mechanism (HIGH-2) - Add ARCHON_CLAUDE_FIRST_EVENT_TIMEOUT_MS and ARCHON_SUPPRESS_NESTED_CLAUDE_WARNING env vars to docs - Add nested Claude Code hang troubleshooting entry - Fix boot module JSDoc: "CLI and server" → "CLI" only - Fix stripCwdEnv JSDoc: remove stale "override: true" reference - Update .claude/rules/cli.md startup behavior section - Update CLAUDE.md @archon/paths description with new exports Tests added: - Assert controller.signal.aborted on timeout - Handle generator that completes immediately without yielding - Strip distinct keys from different .env files Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * simplify: replace string sentinel with typed error class in withFirstMessageTimeout Replace the '__timeout__' string sentinel used to identify timeout rejections with a dedicated FirstEventTimeoutError class. instanceof checks are more explicit and robust than string comparison on error messages. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: address review findings — dotenv version, docs, server warning, marker strip, tests 1. Align dotenv to ^17 (was ^16, rest of monorepo uses ^17.2.3) 2. Remove incorrect SUBPROCESS_ENV_ALLOWLIST claim from docs — the SDK bypasses the env option and uses process.env directly (coleam00#1097) 3. Add CLAUDECODE=1 warning to server entry point (was only in CLI) 4. Add diagnostic payload content test for withFirstMessageTimeout 5. Integrate coleam00#1097's finding: strip CLAUDECODE + CLAUDE_CODE_* session markers (except auth vars) + NODE_OPTIONS + VSCODE_INSPECTOR_OPTIONS from process.env at entry point. Pattern-matched on CLAUDE_CODE_* prefix rather than hardcoding 6 names, so future Claude Code markers are handled automatically. Auth vars (CLAUDE_CODE_OAUTH_TOKEN, CLAUDE_CODE_USE_BEDROCK, CLAUDE_CODE_USE_VERTEX) are preserved. Root cause per coleam00#1097: the Claude Agent SDK leaks process.env into the spawned child regardless of the explicit env option, so the only way to prevent the nested-session deadlock is to delete the markers from process.env at the entry point. Validation: bun run validate passes, 125 paths tests (6 new marker tests), 60 claude tests (1 new diagnostic test), DATABASE_URL leak verified stripped (target repo .env DATABASE_URL does not affect Archon DB selection). * refactor: remove SUBPROCESS_ENV_ALLOWLIST — trust user config, strip only CWD The allowlist was wrong for a single-developer tool: - It blocked keys the user intentionally set in ~/.archon/.env (ANTHROPIC_API_KEY, AWS_*, CLAUDE_CONFIG_DIR, MiniMax vars, etc.) - It was bypassed by the SDK anyway (process.env leaks to subprocess regardless of the env option — see coleam00#1097) - It attracted a constant stream of PRs adding keys (coleam00#1060, coleam00#1093, coleam00#1099) New model: CWD .env keys are the only untrusted source. stripCwdEnv() at entry point handles that. Everything in ~/.archon/.env + shell env passes through to the subprocess. No filtering, no second-guessing. Changes: - Delete env-allowlist.ts and env-allowlist.test.ts - Simplify buildSubprocessEnv() to return { ...process.env } with auth-mode logging (no token stripping — user controls their config) - Replace 4 allowlist-based tests with 1 pass-through test - Remove env-allowlist.test.ts from core test batch - Update security.md and cli.md docs to reflect the new model The CLAUDECODE + CLAUDE_CODE_* marker strip and NODE_OPTIONS strip remain in stripCwdEnv() at entry point — those are process-level safety (not per-subprocess filtering) and are needed regardless. * fix: restore override:true for archon env, add integration tests The integration tests caught a real issue: without override:true, the ~/.archon/.env load doesn't win over shell-inherited env vars. If the user's shell profile exports PORT=9999 and ~/.archon/.env has PORT=3000, the user expects Archon to use 3000. stripCwdEnv() handles CWD .env files (untrusted). override:true handles shell-inherited vars (trusted but less specific than ~/.archon/.env). Different concerns, both needed. Also adds 6 integration tests covering the full entry-point flow: 1. Global auth user with ANTHROPIC_API_KEY in CWD .env — stripped 2. OAuth token in archon env + random key in CWD — CWD stripped, archon kept 3. General leak test — nothing from CWD reaches subprocess 4. Same key in both CWD and archon — archon value wins 5. CLAUDECODE markers stripped even when not from CWD .env 6. CLAUDE_CODE_OAUTH_TOKEN survives marker strip * test: add DATABASE_URL leak scenarios to env integration tests * fix: move CLAUDECODE warning into stripCwdEnv, remove dead useGlobalAuth logic Review findings addressed: 1. CLAUDECODE warning was dead code — the boot import deleted CLAUDECODE from process.env before the warning check in cli.ts/server/index.ts could fire. Moved the warning into stripCwdEnv() itself, emitted BEFORE the deletion. Removed duplicate warning code from both entry points. 2. useGlobalAuth token stripping removed (intentional, not regression) — the old code stripped CLAUDE_CODE_OAUTH_TOKEN and CLAUDE_API_KEY when useGlobalAuth=true. Per design discussion: the user controls ~/.archon/.env and all keys they set are intentional. If they want global auth, they just don't set tokens. Simplified buildSubprocessEnv to log auth mode for diagnostics only, no filtering. 3. Docs "no override needed" corrected — cli.md and configuration.md now reflect the actual code (override: true). --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com> Co-authored-by: Rasmus Widing <rasmus.widing@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Two blockers hit during a fresh setup on macOS, from inside a Claude Code terminal on Bun 1.3.12 — both reproduce deterministically and leave the orchestrator silently hung with no actionable error.
1. Claude Agent SDK subprocess refuses to launch (nested-session guard)
The parent Claude Code shell exports these markers:
Archon's existing cleanup in `buildSubprocessEnv()` (`packages/core/src/clients/claude.ts:167`) only stripped `CLAUDECODE`. Worse, the Claude Agent SDK leaks `process.env` into the spawned child regardless of the explicit `env` option passed to `query()` — so even vars that `SUBPROCESS_ENV_ALLOWLIST` correctly excludes still reach the subprocess. The child then hits its own nested-session guard and the orchestrator hangs after logging `using_global_auth` twice, with no `query_failed`, no `subprocess_error`, no outbound connection to api.anthropic.com, and no visible child process.
The only bulletproof fix until the SDK stops leaking `process.env` is to delete the markers at the process level at server and CLI entry points. Auth vars (`CLAUDE_CODE_OAUTH_TOKEN`, `CLAUDE_CODE_USE_BEDROCK`, `CLAUDE_CODE_USE_VERTEX`) are kept — only the six nested-session markers are removed.
2. Telegram adapter dies on first polling error on Bun 1.3+
Telegraf's `redactToken()` (`node_modules/telegraf/lib/core/network/client.js:248`) does:
```js
error.message = error.message.replace(//(bot|user)(\d+):[^/]+//, '/$1$2:[REDACTED]/');
```
On Bun 1.3+, `Error.message` on certain error instances (fetch/network errors) is non-writable, so the assignment throws `TypeError: Attempted to assign to readonly property.`, kills the polling loop, and leaves the adapter silently dead. The single log line is:
```
telegram.start_failed TypeError: Attempted to assign to readonly property.
at redactToken (telegraf/lib/core/network/client.js:248:5)
```
Patched via `bun patch telegraf` to wrap the assignment in `try/catch`. The original error is still re-thrown unchanged, so Telegraf's own retry/reconnect logic runs normally; the only difference is that we survive one specific Bun runtime quirk.
3. Incidental: `bun.lock` re-sync
`bun patch --commit` re-serialized `bun.lock` and picked up workspace version bumps that were already in the individual `package.json` files (`0.3.5`) but stale in the lockfile (`0.1.0`/`0.2.0`). No behavioral change — just brings the lockfile in sync with reality.
Files
Reproduction
From inside a Claude Code terminal on Bun 1.3+:
```bash
Before this PR
archon chat "say hi"
Hangs forever after "using_global_auth". No error.
After this PR
archon chat "say hi"
"Hello!" (or similar)
```
And for the Telegram bug: configure a bot token, start the server, run `curl https://api.telegram.org/bot/getUpdates` to trigger a 409 conflict in the long-poll — before this PR the adapter logs `telegram.start_failed` and stops polling forever; after, it logs the upstream error and reconnects.
Test plan
Related latent issues (not fixed in this PR)
Summary by CodeRabbit
Bug Fixes
Chores