CodeHarbor

CodeHarbor is a self-hosted Matrix bot and AI chat bridge for codex, claude, and gemini CLI. Users send messages in Matrix rooms, CodeHarbor routes each request to the selected backend, keeps room/session state in SQLite, and sends the final result back to the same room.

Maintainer: https://github.com/biglone Verified by CI:

Quick feedback:

Questions/usage: https://github.com/biglone/CodeHarbor/discussions
Bug report: https://github.com/biglone/CodeHarbor/issues/new?template=bug_report.yml
Feature request: https://github.com/biglone/CodeHarbor/issues/new?template=feature_request.yml
Release announcements (EN/中文 in one thread): #3
Roadmap poll (EN/中文 in one thread): #4
Latest release notes: docs/releases/v0.1.105-release-notes.md
Latest bilingual announcement: docs/releases/v0.1.105-announcement-bilingual.md

Release Notes Index

v0.1.105 notes: docs/releases/v0.1.105-release-notes.md
v0.1.105 announcement (EN/中文): docs/releases/v0.1.105-announcement-bilingual.md
v0.1.104 notes: docs/releases/v0.1.104-release-notes.md
v0.1.104 announcement (EN/中文): docs/releases/v0.1.104-announcement-bilingual.md
v0.1.103 notes: docs/releases/v0.1.103-release-notes.md
v0.1.103 announcement (EN/中文): docs/releases/v0.1.103-announcement-bilingual.md
v0.1.102 notes: docs/releases/v0.1.102-release-notes.md
v0.1.102 announcement (EN/中文): docs/releases/v0.1.102-announcement-bilingual.md
v0.1.69 notes: docs/releases/v0.1.69-release-notes.md
v0.1.69 announcement (EN/中文): docs/releases/v0.1.69-announcement-bilingual.md

What It Does

Matrix channel adapter (receive + reply)
Session-to-backend mapping via persistent SQLite state
One-time migration import from legacy state.json
Duplicate Matrix event protection
Context-aware trigger (DM direct chat + group mention/reply + active session window)
Room-level trigger policy overrides
Runtime backend switch: /backend codex|claude|gemini [model] | /backend auto|status
Cross-backend context bridge on next request after switch
Real /stop cancellation (kills in-flight AI CLI process)
Session runtime workers (logical worker per channel:room:user, with worker stats in /status)
Rate limiting + concurrency guardrails (user/room/global)
Progress + typing updates with group notice coalescing (m.replace edit)
CLI-compat mode (cli_compat_mode) for minimal prompt rewriting + raw event passthrough
Image attachment metadata passthrough from Matrix events into prompt context
Voice attachment transcription (local Whisper/OpenAI fallback)
Semantic file-delivery intent (for example: “把生成的 xxx 文件发送给我”)
Request observability (request_id, queue/exec/send durations, status counters)
NPM-distributed CLI (codeharbor)

Architecture

Matrix Room -> MatrixChannel -> Orchestrator -> AI CLI Executor (codex/claude/gemini)
                                          |
                                          -> StateStore (SQLite)

Implementation Status

Primary runtime: TypeScript/Node (src/, dist/, npm run ...)
Legacy/reference implementation: Python (app/, tests/)
New features and fixes target the TypeScript runtime.
Python code is kept as legacy reference only (maintenance mode).

Prerequisites

Node.js 22+
AI CLI installed and authenticated:
- Codex: codex login
- Claude Code: claude login
- Gemini CLI: gemini
A Matrix bot user + access token

Install

Install globally from npm (after publish):

npm install -g codeharbor

Recommended npm package lifecycle (install/upgrade/verify):

# install latest
npm install -g codeharbor@latest

# verify installed version
codeharbor --version

# upgrade helper (installs latest + restart strategy)
codeharbor self-update

Systemd multi-instance example (Linux):

codeharbor service install --instance bot-a --runtime-home /srv/codeharbor-bot-a --with-admin
codeharbor service install --instance bot-b --runtime-home /srv/codeharbor-bot-b --with-admin

Linux one-command install (creates /opt/codeharbor, sets ownership, installs latest package):

curl -fsSL https://raw.githubusercontent.com/biglone/CodeHarbor/main/scripts/install-linux.sh | bash

Linux easy mode (install + write .env + enable/start systemd in one run):

curl -fsSL https://raw.githubusercontent.com/biglone/CodeHarbor/main/scripts/install-linux-easy.sh | bash -s -- \
  --matrix-homeserver https://matrix.example.com \
  --matrix-user-id @bot:example.com \
  --matrix-access-token 'your-token'

Install first, then enable systemd service with one command:

codeharbor service install

Install + enable main and admin services:

codeharbor service install --with-admin

Restart installed service(s):

codeharbor service restart --with-admin

Remove installed services:

codeharbor service uninstall --with-admin

Notes:

Service commands auto-elevate with sudo when root privileges are required.
codeharbor service install --with-admin and install-linux-easy.sh --enable-admin-service now install /etc/sudoers.d/codeharbor-restart for non-root service users, so Admin UI restart actions work out-of-box.
npm install -g codeharbor@latest now performs best-effort restart per platform:
- Linux: active codeharbor*.service systemd units (including multi-instance units)
- macOS: configured launchd labels (CODEHARBOR_LAUNCHD_MAIN_LABEL, CODEHARBOR_LAUNCHD_ADMIN_LABEL)
- Windows: safe fallback (prints manual PowerShell restart commands)
- set CODEHARBOR_SKIP_POSTINSTALL_RESTART=1 to disable postinstall restart attempts.
If your environment blocks interactive sudo, use explicit fallback:
- sudo <node-bin> <codeharbor-cli-script> service ...

Enable Admin service at install time:

curl -fsSL https://raw.githubusercontent.com/biglone/CodeHarbor/main/scripts/install-linux-easy.sh | bash -s -- \
  --matrix-homeserver https://matrix.example.com \
  --matrix-user-id @bot:example.com \
  --matrix-access-token 'your-token' \
  --enable-admin-service \
  --admin-token 'replace-with-strong-token'

Run local script with custom options:

./scripts/install-linux.sh --app-dir /srv/codeharbor --package codeharbor@0.1.1 --init

Runtime home behavior:

By default, all codeharbor commands use ~/.codeharbor for .env and relative data paths.
Backward compatibility: if /opt/codeharbor/.env already exists, it continues to be used automatically.
No manual cd is required after installation.
To use a custom runtime directory, set CODEHARBOR_HOME (for example export CODEHARBOR_HOME=/srv/codeharbor).

Install directly from GitHub:

npm install -g github:biglone/CodeHarbor

Build a local package tarball and install it:

npm pack
npm install -g ./codeharbor-<version>.tgz

CLI Help Quick Reference

Show CLI help:

codeharbor --help
codeharbor admin --help
codeharbor config --help
codeharbor service --help

Common in-chat control commands:

/help show command help
if Matrix client intercepts slash commands, use escaped //... form for all slash controls (for example //status, //version, //diag queue 5, //diag limiter 5, //upgrade, //autodev init StrawBerry, //autodev run T6.2)
/status show session status, version/update hint, latest upgrade result + recent upgrade ids + upgrade metrics/lock, and runtime metrics
/version force-refresh latest version check
/diag version show runtime version diagnostics (pid/start time/bin path/backend)
/diag media [count] show multimodal diagnostics (image/audio counters + recent records)
/diag upgrade [count] show upgrade diagnostics (distributed lock, aggregate stats, recent upgrade records)
/diag route [count] show backend routing diagnostics (rule hit/fallback reason + recent route records)
/diag autodev [count] show AutoDev diagnostics (stage trace, live loop snapshot, and recent git commit records)
/diag queue [count] show recoverable queue diagnostics (pending/running/retry/failure archive)
/diag limiter [count] show limiter diagnostics (shared mode + denial ratio + recovery latency + recent decision records)
/trace <requestId|latest> show one-request trace (prompt/progress/reply + related workflow/media events; latest resolves the newest request in current session)
Chat final reply includes requestId footer so you can copy it directly for /trace.
- access is restricted to the same session sender or Matrix admin user
/upgrade [version] run self-update and auto-restart service from Matrix DM only
- auth priority: MATRIX_UPGRADE_ALLOWED_USERS > MATRIX_ADMIN_USERS > any DM user (when both empty)
- supports Linux systemd signal restart fallback, macOS launchd/manual fallback, and Windows safe manual fallback
- emits structured success/failure summary with rollback and restart command templates
/backend codex|claude|gemini [model] | /backend auto|status switch or inspect active AI backend (auto restores rule-based routing)
/reset clear current conversation context and suppress one-shot history bridge on the next request
/stop cancel current running request (or queue the stop), clear session context, and drop pending queued tasks for this session
- aliases: /cancel, /esc, /撤回, /撤销

GitHub CI/CD Publish

CodeHarbor supports auto publish to npm from GitHub Actions.

Setup once:

Configure npm Trusted Publishing for this repository/workflow (preferred):
- npm package settings -> Trusted publishing -> Add publisher
- Provider: GitHub Actions
- Repository: biglone/CodeHarbor
- Workflow file: .github/workflows/release-npm.yml
Optional fallback: set repository secret NPM_TOKEN (npm automation token).
Push to main with a publish trigger commit message.

Trigger rules:

push to main + commit message includes [publish-npm] -> run publish workflow
push to main + commit message includes both release and a semver version -> run publish workflow
- examples: release v0.1.1, chore: release 0.1.2
workflow_dispatch -> manual publish from GitHub Actions UI

The workflow runs the same checks as npm run release:verify (changelog + docs guards, typecheck, lint, unit tests, coverage, Admin UI Playwright, build, CLI smoke, package dry-run), then publishes with:

npm publish --provenance --access public

Auth mode selection:

If NPM_TOKEN secret exists: publish with token.
If NPM_TOKEN is absent: publish via npm Trusted Publishing (OIDC).

If the same package version already exists on npm, publish is skipped automatically and the workflow prints a suggested next patch version.

Release checklist (recommended):

Update CHANGELOG.md with a new version section and bullet-point release notes.
Update version in package.json (npm version patch|minor|major).
Run release verification:
- npm run release:verify
Push to main with [publish-npm] or release vX.Y.Z in commit message.
Verify workflow status in GitHub Actions.
Verify package on npm:

npm view codeharbor version

If release CI fails before npm publish, keep the same version and retry publish after fixing CI. Do not skip to the next version number.

Run e2e locally:

npm run test:e2e

If your machine has no system Chrome, install Playwright Chromium once. The test config will then automatically fall back to the bundled browser:

npm run e2e:install
npm run test:e2e

Planning Docs

REQUIREMENTS.md: current baseline + next-stage requirements
TASK_LIST.md: implementation task breakdown and status
docs/USER_MANUAL_ZH.md: Chinese user manual (installation, configuration, verification)
docs/COMPLETE_CONFIGURATION_GUIDE.md: end-to-end setup flow + full feature-to-config mapping
docs/CONFIG_UI_DESIGN.md: configuration UI MVP design
docs/CONFIG_CATALOG.md: consolidated configuration matrix (required/runtime/UI/effective timing)
docs/MULTIMODAL_VERIFICATION_ZH.md: multimodal verification playbook (Codex/Claude image + audio transcription)
docs/GROWTH_PLAYBOOK_ZH.md: growth and community feedback playbook
docs/SOCIAL_PREVIEW_UPLOAD_ZH.md: GitHub social preview image upload guide
docs/DISCUSSION_TEMPLATE_BILINGUAL.md: single-thread bilingual discussion template
docs/ADMIN_STANDALONE_DEPLOY.md: standalone admin deployment and Cloudflare Tunnel exposure guide
docs/BACKUP_AUTOMATION.md: scheduled config backup and restore operations
docs/RELEASE.md: release process and CI/publish policy

Quick Start

For local development from source:

Install dependencies:

npm install

Configure environment:

export CODEHARBOR_HOME="$(pwd)"
codeharbor init

Required values:

MATRIX_HOMESERVER
MATRIX_USER_ID
MATRIX_ACCESS_TOKEN

Run in dev mode:

export CODEHARBOR_HOME="$(pwd)"
npm run dev

Build and run as CLI:

npm run build
export CODEHARBOR_HOME="$(pwd)"
node dist/cli.js start

Configuration Baseline

Use this layered reference to avoid mixing boot-only and runtime tuning items:

It documents:

which keys are required vs optional
which keys can be edited in Admin UI
whether changes are immediate or restart-scoped
hot-update rollback paths (quick hot rollback vs full snapshot rollback)
recommended profiles for local/internal/public deployment
a complete setup sequence from install to production operations

Commands

codeharbor init: guided setup for .env (supports --force to overwrite directly)
codeharbor start: start service
codeharbor doctor: check AI CLI and Matrix connectivity
codeharbor admin serve: start admin UI + config API server
codeharbor service install: install/enable systemd unit(s) after npm install (supports --with-admin)
codeharbor service restart: restart installed systemd unit(s) (supports --with-admin)
codeharbor service uninstall: remove installed systemd unit(s) (supports --with-admin)
codeharbor config export: export current config snapshot as JSON
codeharbor config import <file>: import config snapshot JSON (supports --dry-run)
npm run changelog:check: validate CHANGELOG.md has notes for current package version
npm run docs:check-consistency: validate README/REQUIREMENTS/TASK_LIST version and Node requirement sync
npm run release:verify: run release-grade local verification before publish
scripts/install-linux.sh: Linux bootstrap installer (creates runtime dir + installs npm package)
scripts/install-linux-easy.sh: one-shot Linux install + config + systemd auto-start
scripts/backup-config.sh: export timestamped snapshot and keep latest N backups
scripts/install-backup-timer.sh: install/update user-level systemd timer for automatic backups
npm run test:e2e: run Admin UI end-to-end tests (Playwright; auto-uses system Chrome when available, otherwise bundled Chromium)

Community and Feedback

Questions and usage help: GitHub Discussions (Q&A)
Feature ideas: GitHub Discussions (Ideas) or Feature request issue template
Bug reports: use Bug report issue template for reproducible diagnostics
Contribution guide: CONTRIBUTING.md
Growth and community playbook: docs/GROWTH_PLAYBOOK_ZH.md

Config Backup Script

Create a timestamped snapshot in backups/config and keep latest 20 by default:

./scripts/backup-config.sh

Custom directory and retention:

./scripts/backup-config.sh --dir /var/backups/codeharbor --keep 30

Install/update automatic backup timer:

./scripts/install-backup-timer.sh --schedule "*-*-* 03:30:00" --dir /var/backups/codeharbor --keep 30

Full guide:

docs/BACKUP_AUTOMATION.md

Admin UI And API

Start server:

codeharbor admin serve

Optional overrides:

codeharbor admin serve --host 127.0.0.1 --port 8787

If you bind Admin to a non-loopback host and both ADMIN_TOKEN and ADMIN_TOKENS_JSON are empty, startup is rejected by default. Explicit bypass exists but is not recommended:

codeharbor admin serve --host 0.0.0.0 --allow-insecure-no-token

Open these UI routes in browser:

/ or /settings/global
/settings/bots
/settings/rooms
/health
/audit

/health now includes a CodeHarbor app row with current version, latest version, and update availability.

Main endpoints:

GET /metrics (Prometheus exposition format)
GET /api/admin/auth/status
GET /api/admin/config/global
GET /api/admin/config/skills
PUT /api/admin/config/global
GET /api/admin/config/rooms
GET /api/admin/config/rooms/:roomId
PUT /api/admin/config/rooms/:roomId
DELETE /api/admin/config/rooms/:roomId
GET /api/admin/bot-profiles
PUT /api/admin/bot-profiles
POST /api/admin/bot-profiles/migrate (single-instance -> primary-bot migration, supports dryRun, force, profileId)
POST /api/admin/bot-profiles/apply (supports dryRun, includeDisabled, instanceIds, retireDefaultSingleInstance)
GET /api/admin/health
GET /api/admin/audit?limit=50&kind=config|operations|all&surface=admin|api|webhook&outcome=allowed|denied|error&actor=...&source=...&action=...&method=GET&pathPrefix=/api/...&reasonContains=...&createdFrom=...&createdTo=...
GET /api/admin/sessions?roomId=...&userId=...&from=...&to=...&limit=50&offset=0
GET /api/admin/sessions/export?roomId=...&userId=...&from=...&to=...&limit=50&offset=0&includeMessages=true
GET /api/admin/sessions/:sessionKey/messages?limit=100
GET /api/admin/history/retention
PUT /api/admin/history/retention
POST /api/admin/history/cleanup
GET /api/admin/history/cleanup/runs?limit=20

GET /api/admin/config/skills returns a read-only snapshot for role-skill management:

effective agentWorkflow.roleSkills config
catalog.availableSkills (builtin + discovered local skills)
catalog.missingAssignments (per-role unresolved skill ids)

When ADMIN_TOKEN or ADMIN_TOKENS_JSON is set, requests must include:

Authorization: Bearer <ADMIN_TOKEN>

Access control options:

ADMIN_TOKEN: require bearer token for /api/admin/* and /metrics
ADMIN_TOKENS_JSON: optional multi-token RBAC list (supports admin/viewer role defaults and custom scopes)
API_TOKEN_SCOPES_JSON: optional API token scope override (JSON array, for example ["tasks.submit.api"] or ["tasks.read.api"])
ADMIN_IP_ALLOWLIST: optional comma-separated client IP whitelist (for example 127.0.0.1,192.168.1.10)
ADMIN_ALLOWED_ORIGINS: optional CORS origin allowlist for browser-based cross-origin admin access
EXTERNAL_INTEGRATION_*: optional outbound callback config for API/webhook lifecycle + ticket sync (queued/executing/retrying/completed/failed)

RBAC behavior:

viewer role defaults: admin.read + metrics.read (broad read compatibility)
admin role defaults: admin.read + metrics.read + admin.write
optional scopes in ADMIN_TOKENS_JSON overrides role defaults for that token (supports patterns like admin.read.audit, admin.write.config.*, *)
API token defaults to broad compatibility scopes tasks.submit + tasks.read; API_TOKEN_SCOPES_JSON can narrow it to submit-only/read-only
legacy broad scopes (admin.read, admin.write, tasks.submit, tasks.read, webhook.ingest) still authorize new fine-grained actions for backward compatibility
for ADMIN_TOKENS_JSON, audit actor is derived from token identity (actor field), not x-admin-actor
Admin UI shows current permission status (role/source) after saving auth

External integration callbacks:

inbound POST /api/webhooks/ci|ticket requests are normalized into externalContext and persisted into queue payloads
when EXTERNAL_INTEGRATION_ENABLED=true, CodeHarbor emits non-blocking lifecycle callbacks with short-timeout retries (EXTERNAL_NOTIFY_WEBHOOK_URL)
optional ticket callback (EXTERNAL_TICKET_WEBHOOK_URL) is emitted only for ticket-source tasks
delivery failures never block task execution; outcomes are written into operation audit logs (source=integration:notify|ticket)

Operation audit behavior:

kind=config (default): configuration revision audit entries
kind=operations: authorization and operation outcomes for Admin/API/Webhook (allowed/denied/error)
kind=all: merged timeline of config + operation audit entries (sorted by latest first)
operation entries support additional filters: actor, source, action, method, pathPrefix, reasonContains, createdFrom, createdTo

Metrics quick check:

curl -H "Authorization: Bearer <ADMIN_TOKEN>" \
  http://127.0.0.1:8787/metrics

AutoDev metrics exported:

codeharbor_autodev_runs_total{outcome="succeeded|failed|cancelled"}
codeharbor_autodev_loop_stops_total{reason="no_task|drained|max_runs|deadline|stop_requested|task_incomplete"}
codeharbor_autodev_tasks_blocked_total

Limiter metrics exported:

codeharbor_rate_limiter_decisions_total{source="local|shared|shared_fallback",outcome="allowed|denied"}
codeharbor_rate_limiter_denied_total{reason="..."}
codeharbor_rate_limiter_rejection_ratio
codeharbor_rate_limiter_shared_mode{mode="local|redis"}
codeharbor_rate_limiter_shared_errors_total
codeharbor_rate_limiter_recovery_last_ms / codeharbor_rate_limiter_recovery_avg_ms

Alerting baseline:

Example Prometheus alert rules: docs/PROMETHEUS_ALERT_RULES_EXAMPLE.yml
Includes:
- high request failure ratio
- high queue wait p95
- recent upgrade failure detection
- AutoDev loop stop/block anomaly signals

Rotate tokens quickly (repository script):

./scripts/rotate-admin-token.sh --target rbac --role admin --actor ops-admin
./scripts/rotate-admin-token.sh --target rbac --role viewer --actor ops-audit
./scripts/rotate-admin-token.sh --target rbac --role viewer --actor ops-audit --scopes admin.read.auth,admin.read.audit

Note: PUT /api/admin/config/global always writes .env; high-frequency whitelist keys hot-apply for new requests, while non-whitelist keys still require restart.

Hot Update Rollback

Backup before change:

codeharbor config export -o backups/pre-hot-update.json

Fast rollback for hot-whitelist keys (restartRequired=false):

write previous value back via Admin UI, or call PUT /api/admin/config/global
confirm response hotAppliedKeys contains expected key(s)

Full rollback for mixed/restart-required changes:

codeharbor config import backups/pre-hot-update.json --dry-run
codeharbor config import backups/pre-hot-update.json
codeharbor service restart

Verify + audit:

GET /api/admin/audit?limit=20
GET /api/admin/health
send one Matrix smoke request in each critical room

Boundary: hot updates affect new requests only; in-flight requests are not rolled back.

Admin UI Quick Walkthrough

Start server: codeharbor admin serve.
Open /settings/global, set Admin Token (if enabled), then click Save Auth.
Open Global Settings -> Skills & Advanced:
- click Refresh SKILL catalog to review builtin/local skill IDs
- check Missing SKILL hint before saving role assignments
Adjust global fields and click Save Global Config (UI shows restart-required warning).
Use Restart Main Service or Restart Main + Admin buttons for one-click restart from Admin UI. If services were installed with --with-admin, restart permissions are auto-configured by installer.
Open /settings/rooms, fill Room ID + Workdir, then Save Room.
Open /health to run connectivity checks (codex + Matrix).
Open /audit to verify config revisions (actor/summary/payload).

Primary Bot Governance (Multi-Instance)

Recommended role split in group rooms:

main-hub: primary bot (isPrimary=true), can enable groupDirectModeEnabled=true for non-@ group messages.
dev-main / review-guard: execution/review bots (isPrimary=false), keep group direct mode disabled to avoid cross-talk.

Migration from legacy single-instance:

Open /settings/bots.
Click Migration Dry-Run (POST /api/admin/bot-profiles/migrate with dryRun=true).
Click Migrate As Primary, then Apply Changes.
Optional: enable Retire default single-instance services to disable codeharbor.service / codeharbor-admin.service.

Safety boundaries:

groupDirectModeEnabled is rejected on non-primary profiles.
Direct mode requires an enabled primary profile (isPrimary=true + enabled=true).
Migration returns actionable errors for conflicting primaries; use force=true only when switching ownership intentionally.

Troubleshooting quick checks:

Error groupDirectModeEnabled requires an enabled primary profile: Set one profile to isPrimary=true and enabled=true, or disable direct mode.
Migration blocked by existing primary: Use another profileId or rerun migration with force=true after confirming primary ownership switch.
Group messages no longer trigger without @: Verify primary profile has groupDirectModeEnabled=true and profile changes were applied.

Standalone Admin Deployment

codeharbor admin serve can run as an independent service on target servers without browser/desktop. Access can come from your local browser through a gateway (for example Cloudflare Tunnel).

See:

docs/ADMIN_STANDALONE_DEPLOY.md

Startup Preflight

Before codeharbor start and codeharbor doctor, CodeHarbor runs a preflight check for:

required Matrix env vars
selected AI CLI binary availability (AI_CLI_PROVIDER + CODEX_BIN)
CODEX_WORKDIR validity
.env presence warning

If any check fails, it prints actionable fix commands (for example codeharbor init).

Message Rules

Direct Message (DM)
- all text messages are processed by default (no prefix required)
Group Room
- when GROUP_DIRECT_MODE_ENABLED=true, all non-empty messages are processed directly (no prefix/mention/reply required)
- processed when any allowed trigger matches:
  - message mentions bot user id
  - message replies to a bot message
  - sender has an active conversation window
  - optional explicit prefix match (MATRIX_COMMAND_PREFIX)
Trigger Policy
- GROUP_DIRECT_MODE_ENABLED controls whether groups bypass trigger matching entirely
- global defaults via GROUP_TRIGGER_ALLOW_*
- per-room overrides via ROOM_TRIGGER_POLICY_JSON
Active Conversation Window
- each accepted request activates the sender's conversation in that room
- activation TTL: SESSION_ACTIVE_WINDOW_MINUTES (default: 20)
Control commands
- /help show command cheat sheet for in-chat controls
- if Matrix intercepts /..., use escaped //... command form for all slash controls (for example //status, //version, //diag queue 5, //diag limiter 5, //trace req-123, //trace latest, //upgrade, //autodev init StrawBerry, //autodev run T6.2)
- /status show session + limiter + metrics + runtime worker status, current version, update hint, latest upgrade result, recent upgrade ids, upgrade metrics/lock, and update checked time (cached by TTL)
- /version show current package version and latest-update hint (force refresh)
- /diag version show runtime diagnostics (pid/start time/binary path/backend)
- /diag media [count] show multimodal diagnostics (image/audio counters + recent records)
- /diag upgrade [count] show distributed lock + aggregate stats + recent upgrade run diagnostics
- /diag route [count] show backend routing diagnostics (rule hit + fallback reason + recent route records)
- /diag autodev [count] show AutoDev diagnostics (stage trace + loop status + recent git commit records + error summary)
- /diag queue [count] show queue diagnostics (counts + pending sessions + failure archive)
- /diag limiter [count] show limiter diagnostics (shared mode + rejection/recovery metrics + recent records)
- /trace <requestId|latest> show per-request trace (prompt/progress/reply + related workflow/media events; same-session sender/admin only; latest = current session latest)
- /upgrade [version] install latest (or specified) npm version and trigger service restart (DM only)
  - auth priority: MATRIX_UPGRADE_ALLOWED_USERS > MATRIX_ADMIN_USERS > any DM user (when both empty)
  - includes service-context signal restart fallback when sudo escalation is unavailable
- /backend codex|claude|gemini [model] | /backend auto|status switch backend AI CLI tool at runtime (auto restores rule routing; next request auto-bridges recent local history)
- /reset clear bound Codex session, keep conversation active, and suppress one-shot history bridge on the next request
- /stop cancel in-flight execution (or queue a pending stop when busy), reset session context, and clear pending queue tasks for current session
  - aliases: /cancel, /esc, /撤回, /撤销
- /agents status show multi-agent workflow status for current session (when enabled)
- /agents run <objective> run Planner -> Executor -> Reviewer workflow (when enabled)
- /autodev status show AutoDev doc/task summary + currentTask/nextTask + run snapshot (when enabled)
- /autodev run [taskId] auto-pick pending task (or run specified task) from TASK_LIST.md (when enabled)
- /autodev stop graceful loop stop: finish current task, then stop before next task
- /autodev reconcile reconcile TASK_LIST.md task states from recent AutoDev run records
- /autodev workdir|wd [path]|status|clear show/set/clear AutoDev workdir override for current session
- /autodev init|i [path] [--from file] initialize REQUIREMENTS.md + TASK_LIST.md + docs/AUTODEV_TASK_COMPASS.md in target project
  - short mobile-friendly flow: //autodev init StrawBerry -> //autodev run
  - omit --from to auto-discover design/spec docs in project; use --from to force one source file
- /autodev skills [on|off|summary|progressive|full|status] control role-skill injection and disclosure mode per session
Natural-language file delivery
- supports intent-style requests such as 把生成的 result.mp4 文件发送给我
- for contextual requests such as 这四个视频发给我 or 把那一批都发给我, bot prefers model-driven structured delivery actions over the latest session artifact batch
- for explicit file names/paths, bot still keeps direct rule-based fallback for safety and compatibility
- bot first resolves against the latest session artifact batch captured from recent AI-generated outputs, then falls back to matching local files under current room workdir
- supports batch requests such as 这四个视频文件以消息的形式发给我 or 把生成的视频都发给我, and sends multiple matched attachments sequentially
- if no file name is specified (for example 把生成的文件发送给我), bot picks the most recently updated file in workdir (with size guardrail)

Version update check controls:

PACKAGE_UPDATE_CHECK_ENABLED=true|false
- enable/disable npm latest-version check used by /status, /version, and Admin health app row
PACKAGE_UPDATE_CHECK_TIMEOUT_MS=3000
- timeout (ms) for npm registry version lookup
PACKAGE_UPDATE_CHECK_TTL_MS=21600000
- cache TTL (ms) for update-check results (/status reads cache; /version forces refresh)
MATRIX_ADMIN_USERS=@admin:example.com,@ops:example.com
- optional Matrix admin list used as /upgrade permission fallback when MATRIX_UPGRADE_ALLOWED_USERS is empty
MATRIX_UPGRADE_ALLOWED_USERS=@admin:example.com,@ops:example.com
- optional explicit /upgrade allowlist (higher priority than MATRIX_ADMIN_USERS)
CODEHARBOR_LAUNCHD_MAIN_LABEL=com.codeharbor.main
CODEHARBOR_LAUNCHD_ADMIN_LABEL=com.codeharbor.admin
- optional launchd labels used by macOS upgrade/postinstall restart flow

CLI update helper:

codeharbor self-update
- install latest npm package and run cross-platform restart strategy (Linux systemd / macOS launchd / Windows manual fallback)
- prints structured result summary with rollback + restart command templates for failure recovery

AI CLI backend controls:

AI_CLI_PROVIDER=codex|claude|gemini
- select runtime backend (codex by default)
CODEX_BIN=<path-or-command>
- executable for selected provider (for example codex / claude / gemini)
CODEX_MODEL=<model>
- optional model override for selected provider
CODEX_EXEC_TIMEOUT_MS
- base timeout for one backend execution (default 600000)
- /agents and /autodev use max(CODEX_EXEC_TIMEOUT_MS, 1800000) per role to reduce long-task timeout loops

Cross-backend context bridge behavior:

CodeHarbor stores recent local user/assistant turns per Matrix session.
After /backend codex|claude|gemini [model] or /backend auto, the next non-command request injects a [conversation_bridge] block so the new backend can continue with recent context.
/reset and /stop explicitly suppress this one-shot bridge on the immediate next request so users can start fresh.
CONTEXT_BRIDGE_HISTORY_LIMIT controls how many recent local turns are considered for bridge assembly.
CONTEXT_BRIDGE_MAX_CHARS controls the max bridge payload length (characters).

Backend/model rule routing:

BACKEND_MODEL_ROUTING_RULES_JSON
- optional JSON array rule engine for automatic backend/model selection per request
- conditions support roomIds / senderIds / taskTypes / directMessage / textIncludes / textRegex
- targets support provider (codex|claude|gemini) and/or model; rules are evaluated by priority (high -> low), then declaration order
- when rule target cannot be instantiated (for example no executorFactory runtime), CodeHarbor falls back to default backend and marks status reason as factory_unavailable

Multi-Agent Workflow (Phase B, Opt-In)

AGENT_WORKFLOW_ENABLED=true
- enable /agents and /autodev workflow commands
AGENT_WORKFLOW_AUTO_REPAIR_MAX_ROUNDS
- reviewer reject loop upper bound (default 1)
AGENT_WORKFLOW_PLAN_CONTEXT_MAX_CHARS
- optional planner-plan context char budget per role prompt (default: unlimited / no truncation)
AGENT_WORKFLOW_OUTPUT_CONTEXT_MAX_CHARS
- optional executor/reviewer output context char budget per role prompt (default: unlimited / no truncation)
AGENT_WORKFLOW_FEEDBACK_CONTEXT_MAX_CHARS
- optional reviewer feedback context char budget per repair prompt (default: unlimited / no truncation)
AGENT_WORKFLOW_ROLE_SKILLS_ENABLED=true|false
- enable/disable Planner/Executor/Reviewer role-skill prompt injection (default true)
AGENT_WORKFLOW_ROLE_SKILLS_MODE=summary|progressive|full
- role-skill disclosure mode (progressive default: summary first round, full in later rounds/repair)
AGENT_WORKFLOW_ROLE_SKILLS_MAX_CHARS
- max chars for injected [role_skills] block (default 2400)
AGENT_WORKFLOW_ROLE_SKILLS_ROOTS
- optional comma-separated local skill roots (default ~/.codex/skills)
AGENT_WORKFLOW_ROLE_SKILLS_ASSIGNMENTS_JSON
- optional role-to-skill mapping override JSON (planner/executor/reviewer -> string[])
- defaults map to install-ready builtin fallback skills; local skills with the same id override builtin entries automatically
- builtin fallback skill prompts are maintained in English for consistent global defaults
- default builtin assignment baseline:
  - planner: task-planner, requirements-doc, builtin-planner-core, dependency-analyzer
  - executor: autonomous-dev, bug-finder, test-generator, builtin-executor-core, refactoring
  - reviewer: code-reviewer, security-audit, builtin-reviewer-core, changelog-generator, commit-message
- additional builtin fallbacks are also available for assignment override:
  - planning/design: api-designer, superpowers-workflow, brainstorming, planning-with-files
  - execution/testing: performance-optimizer, auto-code-pipeline, migration-helper, tdd-workflow, webapp-testing, ui-ux-pro-max, pptx, ralph-loop
  - review/release: commit-message, code-simplifier, multi-agent-code-review
AUTODEV_LOOP_MAX_RUNS
- max task attempts for one /autodev run loop execution (default 20, 0 = unlimited)
AUTODEV_LOOP_MAX_MINUTES
- max wall-clock minutes for one /autodev run loop execution (default 120, 0 = unlimited)
AUTODEV_AUTO_COMMIT=true|false
- enable/disable AutoDev git auto-commit after reviewer APPROVED (default true)
AUTODEV_GIT_AUTHOR_NAME
- git author name for AutoDev auto-commit/release commit (default CodeHarbor AutoDev; empty falls back to default)
AUTODEV_GIT_AUTHOR_EMAIL
- git author email for AutoDev auto-commit/release commit (default autodev@codeharbor.local; empty falls back to default)
AUTODEV_AUTO_RELEASE_ENABLED=true|false
- enable/disable AutoDev "big feature done -> release commit" flow (default true)
AUTODEV_AUTO_RELEASE_PUSH=true|false
- push release commit automatically after local release commit (default false)
AUTODEV_RUN_ARCHIVE_ENABLED=true|false
- persist each /autodev run execution archive as local JSON (default true)
AUTODEV_RUN_ARCHIVE_DIR
- archive directory (relative to workdir or absolute path, default .codeharbor/autodev-runs)
AUTODEV_STAGE_OUTPUT_ECHO_ENABLED=true|false
- echo planner/executor/reviewer full stage output to Matrix during /autodev run (default true)
AUTODEV_PREFLIGHT_AUTO_STASH=true|false
- when git preflight detects a dirty worktree, auto-stash (git stash --include-untracked) and continue run (default false)
AUTODEV_MAX_CONSECUTIVE_FAILURES
- when the same task fails repeatedly and reaches this threshold, mark it 🚫 blocked (default 3)
AUTODEV_VALIDATION_STRICT=true|false
- fail closed on validation gate when structured evidence is missing (VALIDATION_STATUS/__EXIT_CODES__); default false

AutoDev (/autodev) conventions:

Architecture, control-chain, and troubleshooting handbook: docs/AUTODEV_OPERATIONS_ZH.md.
Workspace must contain REQUIREMENTS.md and TASK_LIST.md.
/autodev init|i [path] [--from file] scaffolds missing AutoDev files and binds workdir override for current session.
- when --from is omitted, CodeHarbor auto-discovers design/spec docs and uses them to generate initial REQUIREMENTS/TASK_LIST templates.
- init uses a 3-stage flow: Stage-A deterministic scaffold, Stage-B AI enhancement, Stage-C hard validation with fallback to Stage-A baseline on invalid output.
- Stage-B AI enhancement runs only when both REQUIREMENTS.md and TASK_LIST.md were generated in this init run (existing files are preserved).
- if path is a project name (no /), CodeHarbor resolves both <room_workdir>/<name> and sibling <parent>/<name>.
- when target path is missing/not-directory, init fails explicitly and keeps current workdir unchanged.
/autodev workdir|wd [path]|status|clear inspects or changes session-level workdir override.
TASK_LIST.md should include task IDs and status markers (⬜, 🔄, ✅, ❌, 🚫) in table rows or checklist rows.
TASK_LIST.md task status is system-managed by orchestrator; avoid manual edits and use /autodev reconcile when drift needs healing.
/autodev run (without task id) loops through task list: selects 🔄 first, then ⬜, and keeps running until no executable task remains.
/autodev run loop is guarded by AUTODEV_LOOP_MAX_RUNS and AUTODEV_LOOP_MAX_MINUTES; reaching either limit pauses safely with a summary notice, and you can resume with /autodev run.
/autodev run <taskId> runs only the specified task.
/autodev stop does not interrupt the current task; it stops loop scheduling after the current task completes.
/autodev reconcile performs one-shot task-state reconciliation using recent AutoDev run records (useful after manual edits or interrupted sessions).
Loop guardrail rules:
- AUTODEV_LOOP_MAX_RUNS=0 means unlimited loop rounds.
- AUTODEV_LOOP_MAX_MINUTES=0 means unlimited loop wall-clock time.
- Hitting run/time limit is a safe pause (not hard failure): AutoDev prints remaining task summary and resume hint.
- Resume command is always /autodev run (or /autodev run <taskId> for targeted rerun).
- /autodev run <taskId> is single-task mode and does not consume loop run/time budgets.
- Recommended for long roadmap execution: set both loop limits to 0, and keep AUTODEV_MAX_CONSECUTIVE_FAILURES + no-progress detection enabled as safety rails.
/autodev skills ... controls role-skill injection (on|off) and disclosure mode (summary|progressive|full) for current session.
/autodev content on|off|status controls AutoDev stage output echo (planner/executor/reviewer content) for current session.
Task closes to ✅ only when completion gate passes (reviewer approved + no explicit validation failure + auto-commit success when commit is required).
In strict mode (AUTODEV_VALIDATION_STRICT=true), completion gate requires structured validation evidence (VALIDATION_STATUS and/or __EXIT_CODES__).
When workflow/reviewer execution succeeds but completion gate fails, /autodev status reports runState: completed_with_gate_failed (instead of succeeded).
/autodev status exposes validation observability fields: runValidationFailureClass, runValidationEvidenceSource, and runValidationAt.
Validation fuse rule: if the same task hits the same validationFailureClass consecutively for AUTODEV_MAX_CONSECUTIVE_FAILURES, AutoDev stops and marks it 🚫 with next-action guidance.
When reviewer verdict is APPROVED and the workdir is a clean Git repo, CodeHarbor auto-commits changes with a semantic subject: <type>(<scope>): <business-summary> (<taskId>).
AutoDev commit intent uses a hybrid strategy: prefer reviewer SUMMARY (role-skill output) when it matches the selected commit language; otherwise fall back to deterministic template inference.
Commit language policy: for a brand-new project (no history) the first AutoDev commit defaults to English; for existing projects, AutoDev follows the recent repository commit language trend.
AutoDev commit body includes Task-ID, Changed-files, and Generated-by for traceability.
AutoDev result notice always includes git commit status and changed files (git changed files).
If TASK_LIST.md has a dedicated 发布映射 section with rows like | T8.4 | v0.1.55 | ... |, AutoDev can create a follow-up release commit after task completion: release: vX.Y.Z [publish-npm] (updates package.json/package-lock.json/CHANGELOG.md).
AutoDev release mapping parser only reads the 发布映射 section; keep community-priority/milestone tables in separate headings (for example 社区优先级 -> 可执行里程碑) to avoid ambiguous version parsing.
Current community roadmap milestones are aligned as: T8.1~T8.3 -> v0.1.53~v0.1.54, T8.4~T8.5 -> v0.1.55~v0.1.56, T8.6 -> v0.1.57, T8.7~T8.8 -> v0.1.58~v0.1.59.
When CI detects that the target version already exists on npm, the release workflow skips publishing and prints Suggested next version to keep release flow idempotent.
If the same task fails consecutively and reaches AUTODEV_MAX_CONSECUTIVE_FAILURES, CodeHarbor marks it as 🚫 and skips it in later loops.
If the repo is missing or already dirty before run, AutoDev skips commit and reports the reason in the result notice.
When AUTODEV_PREFLIGHT_AUTO_STASH=true, dirty git preflight is auto-stashed and the run continues (stash ref is reported in notices).
If workflow attempts to edit TASK_LIST.md, AutoDev rolls it back and fails completion gate with task-list-policy-violated.
When using scripts/autodev-loop-runner.sh, a new trigger is skipped while any task is already 🔄 in progress.

Default is disabled to keep legacy behavior unchanged.

CLI Compatibility Mode

To make IM behavior closer to local codex CLI interaction, enable:

CLI_COMPAT_MODE=true
- preserve user prompt whitespace
- avoid stripping @bot mention text in prompt body
- enable richer raw event passthrough summaries
CLI_COMPAT_PASSTHROUGH_EVENTS=true
- emit raw event summaries from codex JSON stream
CLI_COMPAT_PRESERVE_WHITESPACE=true
- keep incoming Matrix message body untrimmed for execution
CLI_COMPAT_DISABLE_REPLY_CHUNK_SPLIT=true|false
- optionally send one full message chunk to Matrix without auto split
CLI_COMPAT_PROGRESS_THROTTLE_MS
- lower update throttle for near-real-time progress
CLI_COMPAT_FETCH_MEDIA=true|false
- download Matrix mxc:// media (image) to temp file and pass image context to backend
CLI_COMPAT_IMAGE_MAX_BYTES
- per-image max size guard (bytes), oversized images are skipped with user notice
CLI_COMPAT_IMAGE_MAX_COUNT
- max number of images passed to backend in one request
CLI_COMPAT_IMAGE_ALLOWED_MIME_TYPES
- comma-separated image MIME allowlist (image/png,image/jpeg,...)
CLI_COMPAT_TRANSCRIBE_AUDIO=true|false
- download Matrix m.audio attachments and transcribe them into prompt context
CLI_COMPAT_AUDIO_TRANSCRIBE_MODEL
- OpenAI transcription model (default gpt-4o-mini-transcribe)
CLI_COMPAT_AUDIO_TRANSCRIBE_TIMEOUT_MS
- timeout for each audio transcription request
CLI_COMPAT_AUDIO_TRANSCRIBE_MAX_CHARS
- max transcript length appended to prompt for one attachment
CLI_COMPAT_AUDIO_TRANSCRIBE_MAX_RETRIES
- retry count for local/OpenAI transcription failures (default 1)
CLI_COMPAT_AUDIO_TRANSCRIBE_RETRY_DELAY_MS
- base retry delay between attempts
CLI_COMPAT_AUDIO_TRANSCRIBE_MAX_BYTES
- skip transcription when attachment is larger than this size
CLI_COMPAT_AUDIO_LOCAL_WHISPER_COMMAND
- optional local whisper command template (use {input} placeholder for audio file path)
- helper command shipped by package: codeharbor-whisper-transcribe --input {input} --model small
CLI_COMPAT_AUDIO_LOCAL_WHISPER_TIMEOUT_MS
- timeout for local whisper command execution
CLI_COMPAT_RECORD_PATH=/abs/path/records.jsonl
- append executed prompts as JSONL for replay benchmarking

Note: execution still uses codex exec/resume per request; compatibility mode focuses on behavior parity and reduced middleware interference.

Persistence

STATE_DB_PATH=data/state.db
- SQLite store for sessions + processed event ids
STATE_PATH=data/state.json
- legacy JSON source for one-time migration import when SQLite is empty
MAX_SESSION_AGE_DAYS=30
- prune stale sessions by age
MAX_SESSIONS=5000
- prune least-recently-updated sessions when over limit

Rate Limiting

RATE_LIMIT_WINDOW_SECONDS
RATE_LIMIT_MAX_REQUESTS_PER_USER
RATE_LIMIT_MAX_REQUESTS_PER_ROOM
RATE_LIMIT_MAX_CONCURRENT_GLOBAL
RATE_LIMIT_MAX_CONCURRENT_PER_USER
RATE_LIMIT_MAX_CONCURRENT_PER_ROOM
RATE_LIMIT_SHARED_MODE=local|redis
RATE_LIMIT_SHARED_REDIS_URL (required when shared mode is redis)
RATE_LIMIT_SHARED_REDIS_KEY_PREFIX
RATE_LIMIT_SHARED_REDIS_COMMAND_TIMEOUT_MS
RATE_LIMIT_SHARED_REDIS_CONCURRENCY_TTL_MS
RATE_LIMIT_SHARED_FALLBACK_TO_LOCAL=true|false

Set a value to 0 to disable a specific limiter.

Codex CLI Alignment

Use these to align runtime with your terminal CLI profile:

CODEX_SANDBOX_MODE
CODEX_APPROVAL_POLICY
CODEX_EXTRA_ARGS
CODEX_EXTRA_ENV_JSON

When image attachments are present and CLI_COMPAT_FETCH_MEDIA=true, CodeHarbor will:

download mxc:// media to a temp file
apply image policy guardrails (CLI_COMPAT_IMAGE_MAX_BYTES, CLI_COMPAT_IMAGE_MAX_COUNT, CLI_COMPAT_IMAGE_ALLOWED_MIME_TYPES)
for Codex backend, pass local file paths as --image
for Claude backend, use stream-json input with base64 image blocks
if Claude image input fails, auto-retry once without image blocks and notify user
best-effort cleanup temp files after the request
optional prompt record append (CLI_COMPAT_RECORD_PATH) for deterministic replay input

When audio attachments are present and both CLI_COMPAT_FETCH_MEDIA=true and CLI_COMPAT_TRANSCRIBE_AUDIO=true, CodeHarbor will:

download m.audio media to a temp file
skip oversized audio files based on CLI_COMPAT_AUDIO_TRANSCRIBE_MAX_BYTES
if CLI_COMPAT_AUDIO_LOCAL_WHISPER_COMMAND is configured, execute local whisper first
if local whisper fails and OPENAI_API_KEY is available, fallback to OpenAI transcription API
retry transient failures using CLI_COMPAT_AUDIO_TRANSCRIBE_MAX_RETRIES
append transcript to [audio_transcripts] prompt block
continue request even if transcription fails (warn log + no transcript)
best-effort cleanup temp files after the request

OPENAI_API_KEY is optional when local whisper command is configured, and required only for OpenAI fallback. For codeharbor-whisper-transcribe, install runtime first: python3 -m pip install faster-whisper.

Replay Benchmark

Replay recorded prompts directly against codex CLI to quantify drift and latency:

npm run replay:cli-compat -- --input data/cli-compat-record.jsonl --out data/replay-report.json --max 50

Useful flags:

--model <name>
--workdir <path>
--timeout-ms <n>
--sandbox <mode>
--approval <policy>
--dangerous

Progress + Output

MATRIX_PROGRESS_UPDATES=true
- emit stage progress updates (for example reasoning/thinking snippets)
MATRIX_PROGRESS_MIN_INTERVAL_MS=2500
- minimum interval between progress updates
MATRIX_PROGRESS_DELIVERY_MODE=upsert|timeline
- upsert edits one progress notice in group chats; timeline appends progress notices
MATRIX_TYPING_TIMEOUT_MS=10000
- typing indicator timeout; CodeHarbor refreshes typing state while handling a request
MATRIX_NOTICE_BADGE_ENABLED=true|false
- enable/disable rich-message badge headers (CodeHarbor 提示 / CodeHarbor AI 回复)
Group rooms default to notice edit (m.replace) to coalesce progress and reduce spam (MATRIX_PROGRESS_DELIVERY_MODE=upsert).
Reply chunking is paragraph/code-block aware to avoid cutting fenced blocks when possible.

Tests

npm run typecheck
npm test
npm run build
npm run test:legacy

If Python legacy dependencies are missing, install them first:

python3 -m pip install -r requirements.txt

Legacy Runtime

Legacy Python runtime exists in app/ and tests/.
It is not part of default release/CI gates.
Use npm run test:legacy for optional regression checks.

Name		Name	Last commit message	Last commit date
Latest commit History 531 Commits
.github		.github
app		app
docs		docs
e2e		e2e
scripts		scripts
src		src
test		test
tests		tests
.env.example		.env.example
.gitignore		.gitignore
CHANGELOG.md		CHANGELOG.md
CONTRIBUTING.md		CONTRIBUTING.md
LEGACY.md		LEGACY.md
LICENSE		LICENSE
README.md		README.md
REQUIREMENTS.md		REQUIREMENTS.md
TASK_LIST.md		TASK_LIST.md
eslint.config.mjs		eslint.config.mjs
package-lock.json		package-lock.json
package.json		package.json
playwright.config.ts		playwright.config.ts
requirements.txt		requirements.txt
tsconfig.json		tsconfig.json
vitest.config.ts		vitest.config.ts

Folders and files

Latest commit

History

Repository files navigation

CodeHarbor

Release Notes Index

What It Does

Architecture

Implementation Status

Prerequisites

Install

CLI Help Quick Reference

GitHub CI/CD Publish

Planning Docs

Quick Start

Configuration Baseline

Commands

Community and Feedback

Config Backup Script

Admin UI And API

Hot Update Rollback

Admin UI Quick Walkthrough

Primary Bot Governance (Multi-Instance)

Standalone Admin Deployment

Startup Preflight

Message Rules

Multi-Agent Workflow (Phase B, Opt-In)

CLI Compatibility Mode

Persistence

Rate Limiting

Codex CLI Alignment

Replay Benchmark

Progress + Output

Tests

Legacy Runtime

About

Topics

Resources

License

Contributing

Uh oh!

Stars

Watchers

Forks

Releases 10

Packages 0

Uh oh!

Contributors

Uh oh!

Languages

Packages