Transport Guide

Run all commands from repository root.

Unified operator control-plane snapshot (all core transport/runtime surfaces):

cargo run -p tau-coding-agent -- \
  --operator-control-summary \
  --operator-control-summary-json

Troubleshooting map and field details: docs/guides/operator-control-summary.md.

Runtime Log Rotation

Runtime JSONL logs use size-based rotation with bounded retained files.

TAU_LOG_ROTATION_MAX_BYTES: max bytes for active log before rollover (default 10485760 / 10 MiB).
TAU_LOG_ROTATION_MAX_FILES: retained file count including active file (default 5).

Retained files are named:

Active: <name>.jsonl
Backups: <name>.jsonl.1, <name>.jsonl.2, ...

Examples:

# Keep active + 2 backups, rotate every 2 MiB.
export TAU_LOG_ROTATION_MAX_BYTES=2097152
export TAU_LOG_ROTATION_MAX_FILES=3

Removed Contract Runner Migration Matrix

Removed Flag	Status	Replacement Path
`--memory-contract-runner`	removed	Use built-in memory tools (`memory_write`, `memory_read`, `memory_search`, `memory_tree`) with `--memory-state-dir` and `--transport-health-inspect memory`.
`--dashboard-contract-runner`	removed	Use gateway/API surfaces (`--gateway-openresponses-server`) with dashboard diagnostics (`--dashboard-status-inspect`, `--transport-health-inspect dashboard`).
`--browser-automation-contract-runner`	removed	Use `--browser-automation-live-runner` plus `--browser-automation-live-fixture` and `--browser-automation-playwright-cli`.
`--custom-command-contract-runner`	removed	Preserve `--custom-command-state-dir` data and use `--custom-command-status-inspect` with `--transport-health-inspect custom-command`.

GitHub Issues bridge

export GITHUB_TOKEN=...your-token...

cargo run -p tau-coding-agent -- \
  --model openai/gpt-5.2 \
  --github-issues-bridge \
  --github-repo owner/repo \
  --github-bot-login your-bot-login \
  --github-required-label tau-ready \
  --github-issue-number 7 \
  --github-state-dir .tau/github-issues \
  --github-poll-interval-seconds 30 \
  --github-artifact-retention-days 30

--github-required-label can be repeated; only issues with at least one matching label are processed. --github-issue-number can be repeated; only matching issue numbers are processed.

Run exactly one poll cycle (useful for CI smoke jobs and cron workflows):

cargo run -p tau-coding-agent -- \
  --model openai/gpt-5.2 \
  --github-issues-bridge \
  --github-repo owner/repo \
  --github-poll-once \
  --github-state-dir .tau/github-issues

Bridge control commands in issue comments:

/tau help
/tau status
/tau health
/tau auth <status|matrix> ...
/tau doctor [--online]
/tau stop
/tau chat start|resume|reset|status|summary|replay|show|search|export
/tau artifacts|artifacts run <run_id>|artifacts show <artifact_id>|artifacts purge
/tau demo-index list|run [scenario[,scenario...]] [--timeout-seconds <n>]|report

Issue command response schema (normalized):

Footer format: Tau command `<command>` | status `<status>` | reason_code `<reason_code>` .
Status taxonomy:
acknowledged: command accepted (for example stop/cancel acknowledgement).
reported: command completed and returned diagnostics/control output.
failed: command execution failed.
Artifact pointers: label: id=\...` path=`...` bytes=`...``.
Oversized command output is truncated deterministically and spilled to a channel-store artifact:
output_truncated: true
overflow_artifact: id=\...` path=`...` bytes=`...``

Auth diagnostics commands:

/tau auth status: report provider auth posture with strict-subscription context.
/tau auth matrix: report cross-provider mode/availability matrix and filters.

Doctor diagnostics command:

/tau doctor: run bounded local diagnostics and post summary plus artifact pointers.
/tau doctor --online: include remote release-update lookup (network dependent).

Demo-index commands for issue-driven demos:

/tau demo-index list: show allowlisted scenarios and expected markers.
/tau demo-index run onboarding,gateway-auth --timeout-seconds 120: execute bounded demo scenarios from the issue thread and persist report/log artifacts.
/tau demo-index report: show latest demo-index report artifact pointers for the issue channel.

Inspect deterministic GitHub bridge state/report output:

cargo run -p tau-coding-agent -- \
  --github-status-inspect owner/repo \
  --github-state-dir .tau/github-issues \
  --github-status-json

Slack Socket Mode bridge

export TAU_SLACK_APP_TOKEN=...xapp-token...
export TAU_SLACK_BOT_TOKEN=...xoxb-token...

cargo run -p tau-coding-agent -- \
  --model openai/gpt-5.2 \
  --slack-bridge \
  --slack-state-dir .tau/slack \
  --slack-artifact-retention-days 30 \
  --slack-thread-detail-output true \
  --slack-thread-detail-threshold-chars 1500

Multi-channel contract runner (Telegram, Discord, WhatsApp)

Use this fixture-driven runtime mode to validate channel-store writes, retry behavior, and deduplication for supported transports.

cargo run -p tau-coding-agent -- \
  --model openai/gpt-5.2 \
  --multi-channel-contract-runner \
  --multi-channel-fixture crates/tau-multi-channel/testdata/multi-channel-contract/baseline-three-channel.json \
  --multi-channel-state-dir .tau/multi-channel \
  --multi-channel-queue-limit 64 \
  --multi-channel-processed-event-cap 10000 \
  --multi-channel-retry-max-attempts 4 \
  --multi-channel-retry-base-delay-ms 0 \
  --multi-channel-media-understanding true \
  --multi-channel-media-max-attachments 4 \
  --multi-channel-media-max-summary-chars 280

The runner writes channel-store output under:

.tau/multi-channel/channel-store/telegram/...
.tau/multi-channel/channel-store/discord/...
.tau/multi-channel/channel-store/whatsapp/...

Runtime state for duplicate suppression is persisted at:

.tau/multi-channel/state.json
.tau/multi-channel/runtime-events.jsonl (per-cycle observability log)

Inbound channel-store logs also include per-event media_understanding diagnostics for supported image/audio/video attachments (and explicit skip/failure reason codes for unsupported media).

Inspect multi-channel transport health snapshot:

cargo run -p tau-coding-agent -- \
  --multi-channel-state-dir .tau/multi-channel \
  --transport-health-inspect multi-channel \
  --transport-health-json

Inspect multi-channel rollout guardrail/status report:

cargo run -p tau-coding-agent -- \
  --multi-channel-state-dir .tau/multi-channel \
  --multi-channel-status-inspect \
  --multi-channel-status-json

Operational rollout and rollback guidance: docs/guides/multi-channel-ops.md.

Multi-channel live runner (Telegram, Discord, WhatsApp)

Use this deterministic live-ingress mode to process local adapter inbox files without external provider calls.

Ingress directory layout:

.tau/multi-channel/live-ingress/telegram.ndjson
.tau/multi-channel/live-ingress/discord.ndjson
.tau/multi-channel/live-ingress/whatsapp.ndjson

Each line is one normalized provider envelope JSON object.

Run readiness preflight before enabling live mode (fails closed on required gaps):

cargo run -p tau-coding-agent -- \
  --multi-channel-live-readiness-preflight

JSON output mode:

cargo run -p tau-coding-agent -- \
  --multi-channel-live-readiness-preflight \
  --multi-channel-live-readiness-json

Readiness checks cover:

Credential store readability (--credential-store)
Live ingress directory path + per-channel inbox files
Channel prerequisites for Telegram/Discord/WhatsApp via env var or integration secret (/integration-auth set <integration-id> <secret>)

Required channel secrets:

Telegram: TAU_TELEGRAM_BOT_TOKEN or integration id telegram-bot-token
Discord: TAU_DISCORD_BOT_TOKEN or integration id discord-bot-token
WhatsApp access token: TAU_WHATSAPP_ACCESS_TOKEN or integration id whatsapp-access-token
WhatsApp phone number id: TAU_WHATSAPP_PHONE_NUMBER_ID or integration id whatsapp-phone-number-id

One-shot provider payload ingest (raw payload -> live ingress NDJSON):

cargo run -p tau-coding-agent -- \
  --multi-channel-live-ingest-file ./crates/tau-multi-channel/testdata/multi-channel-live-ingress/raw/telegram-update.json \
  --multi-channel-live-ingest-transport telegram \
  --multi-channel-live-ingest-provider telegram-bot-api \
  --multi-channel-live-ingest-dir .tau/multi-channel/live-ingress

Supported ingest transports:

telegram
discord
whatsapp

The command validates payload shape and appends one normalized envelope line to:

<ingest-dir>/telegram.ndjson
<ingest-dir>/discord.ndjson
<ingest-dir>/whatsapp.ndjson

cargo run -p tau-coding-agent -- \
  --model openai/gpt-5.2 \
  --multi-channel-live-runner \
  --multi-channel-live-ingress-dir .tau/multi-channel/live-ingress \
  --multi-channel-state-dir .tau/multi-channel \
  --multi-channel-queue-limit 64 \
  --multi-channel-processed-event-cap 10000 \
  --multi-channel-retry-max-attempts 4 \
  --multi-channel-retry-base-delay-ms 0

The live runner writes to the same state and channel-store paths as contract mode:

.tau/multi-channel/state.json
.tau/multi-channel/runtime-events.jsonl
.tau/multi-channel/channel-store/<transport>/<channel>/...

Policy controls for DM/group behavior and mention gating:

.tau/security/channel-policy.json (dmPolicy, allowFrom, groupPolicy, requireMention)
.tau/security/allowlist.json and .tau/security/pairings.json (actor access controls)

Multi-agent contract runner

Use this fixture-driven runtime mode to validate planner/delegated/review route selection, retry handling, deduplication, and routed-case snapshot persistence.

cargo run -p tau-coding-agent -- \
  --model openai/gpt-5.2 \
  --multi-agent-contract-runner \
  --multi-agent-fixture crates/tau-coding-agent/testdata/multi-agent-contract/rollout-pass.json \
  --multi-agent-state-dir .tau/multi-agent \
  --multi-agent-queue-limit 64 \
  --multi-agent-processed-case-cap 10000 \
  --multi-agent-retry-max-attempts 4 \
  --multi-agent-retry-base-delay-ms 0

The runner writes state and observability output under:

.tau/multi-agent/state.json
.tau/multi-agent/runtime-events.jsonl
.tau/multi-agent/channel-store/multi-agent/orchestrator-router/...

Inspect multi-agent transport health snapshot:

cargo run -p tau-coding-agent -- \
  --multi-agent-state-dir .tau/multi-agent \
  --transport-health-inspect multi-agent \
  --transport-health-json

Inspect multi-agent rollout guardrail/status report:

cargo run -p tau-coding-agent -- \
  --multi-agent-state-dir .tau/multi-agent \
  --multi-agent-status-inspect \
  --multi-agent-status-json

Operational rollout and rollback guidance: docs/guides/multi-agent-ops.md.

Semantic memory diagnostics (contract runner removed)

--memory-contract-runner is removed. Runtime memory ownership is in tau-agent-core (auto recall, retrieval ranking, and embedding-backed matching). tau-memory provides the shared file-backed memory store/runtime helpers plus fixture schemas for contract validation. Built-in tools (memory_write, memory_read, memory_search, memory_tree) operate against --memory-state-dir.

Use transport-health inspection for deterministic operator diagnostics against persisted memory health artifacts.

Inspect semantic memory health snapshot:

cargo run -p tau-coding-agent -- \
  --memory-state-dir .tau/memory \
  --transport-health-inspect memory \
  --transport-health-json

Operational migration and diagnostics guidance: docs/guides/memory-ops.md.

Browser automation live runner

cargo run -p tau-coding-agent -- \
  --model openai/gpt-5.2 \
  --browser-automation-live-runner \
  --browser-automation-live-fixture crates/tau-coding-agent/testdata/browser-automation-live/live-sequence.json \
  --browser-automation-playwright-cli playwright-cli \
  --browser-automation-state-dir .tau/browser-automation \

The live runner executes fixture cases through an external Playwright-compatible CLI and writes state and observability output under:

.tau/browser-automation/state.json
.tau/browser-automation/runtime-events.jsonl

Inspect browser automation transport health snapshot:

cargo run -p tau-coding-agent -- \
  --browser-automation-state-dir .tau/browser-automation \
  --transport-health-inspect browser-automation \
  --transport-health-json

Run browser automation readiness preflight:

cargo run -p tau-coding-agent -- \
  --browser-automation-preflight

JSON output mode:

cargo run -p tau-coding-agent -- \
  --browser-automation-preflight \
  --browser-automation-preflight-json

Troubleshooting:

browser_automation.npx not ready: install Node.js/npm and ensure npx is on PATH.
browser_automation.playwright_cli missing: install @playwright/mcp or set --browser-automation-playwright-cli to a valid wrapper binary.
--browser-automation-contract-runner has been removed; use the live-runner command shown above.

Demo command path:

./scripts/demo/browser-automation.sh
./scripts/demo/all.sh --only browser-automation --fail-fast

Dashboard contract runner (removed)

--dashboard-contract-runner has been removed from active transport dispatch. Use gateway/API surfaces plus diagnostics commands against persisted state artifacts.

Diagnostics commands read state and observability output under:

.tau/dashboard/state.json
.tau/dashboard/runtime-events.jsonl
.tau/dashboard/channel-store/dashboard/<channel_id>/...

Inspect dashboard transport health snapshot:

cargo run -p tau-coding-agent -- \
  --dashboard-state-dir .tau/dashboard \
  --transport-health-inspect dashboard \
  --transport-health-json

Inspect dashboard rollout guardrail/status report:

cargo run -p tau-coding-agent -- \
  --dashboard-state-dir .tau/dashboard \
  --dashboard-status-inspect \
  --dashboard-status-json

Operational rollout and rollback guidance: docs/guides/dashboard-ops.md.

Gateway OpenResponses + websocket control server

Run the authenticated gateway server for HTTP OpenResponses plus websocket control-plane methods.

cargo run -p tau-coding-agent -- \
  --model openai/gpt-5.2 \
  --gateway-openresponses-server \
  --gateway-openresponses-bind 127.0.0.1:8787 \
  --gateway-openresponses-auth-mode token \
  --gateway-openresponses-auth-token dev-secret \
  --gateway-state-dir .tau/gateway \
  --gateway-openresponses-rate-limit-window-seconds 60 \
  --gateway-openresponses-rate-limit-max-requests 120

Server endpoints:

POST /v1/responses (OpenResponses subset)
POST /v1/chat/completions (OpenAI-compatible chat surface)
POST /v1/completions (OpenAI-compatible completions surface)
GET /v1/models (OpenAI-compatible model listing)
GET /gateway/sessions
GET /gateway/sessions/{session_key}
POST /gateway/sessions/{session_key}/append (requires policy_gate=allow_session_write)
POST /gateway/sessions/{session_key}/reset (requires policy_gate=allow_session_write)
GET /gateway/memory/{session_key}
PUT /gateway/memory/{session_key} (requires policy_gate=allow_memory_write)
POST /gateway/ui/telemetry
POST /gateway/auth/session (only when --gateway-openresponses-auth-mode=password-session)
GET /gateway/status
GET /gateway/ws (websocket control protocol)
GET /webchat

Websocket control methods (schema versions 0 and 1 accepted):

capabilities.request
gateway.status.request
session.status.request
session.reset.request
run.lifecycle.status.request

Example websocket frame:

{"schema_version":1,"request_id":"req-cap","kind":"capabilities.request","payload":{}}

The server returns deterministic response envelopes and error frames for malformed or unsupported input. It also emits heartbeat signals (ws.ping plus gateway.heartbeat) on a fixed interval.

Compatibility fixtures for protocol replay:

crates/tau-coding-agent/testdata/gateway-ws-protocol/dispatch-supported-controls.json
crates/tau-coding-agent/testdata/gateway-ws-protocol/dispatch-unsupported-schema-continues.json
crates/tau-coding-agent/testdata/gateway-ws-protocol/dispatch-unknown-kind-continues.json

Tau daemon lifecycle

Use daemon lifecycle commands to install/start/stop/status/uninstall Tau daemon state and profile files.

Install profile files (auto resolves host profile):

cargo run -p tau-coding-agent -- \
  --daemon-install \
  --daemon-state-dir .tau/daemon \
  --daemon-profile auto

Start and stop lifecycle state:

cargo run -p tau-coding-agent -- \
  --daemon-start \
  --daemon-state-dir .tau/daemon

cargo run -p tau-coding-agent -- \
  --daemon-stop \
  --daemon-stop-reason maintenance_window \
  --daemon-state-dir .tau/daemon

Inspect status and diagnostics:

cargo run -p tau-coding-agent -- \
  --daemon-status \
  --daemon-status-json \
  --daemon-state-dir .tau/daemon

Uninstall profile files:

cargo run -p tau-coding-agent -- \
  --daemon-uninstall \
  --daemon-state-dir .tau/daemon

Subcommand alias is also supported:

cargo run -p tau-coding-agent -- daemon status --json --state-dir .tau/daemon

Generated profile files:

launchd: .tau/daemon/launchd/io.tau.coding-agent.plist
systemd user: .tau/daemon/systemd/tau-coding-agent.service

Runbook and troubleshooting commands: docs/guides/daemon-ops.md.

Gateway contract runner

Use this fixture-driven runtime mode to validate Tau gateway request handling, retry outcomes, state persistence, and channel-store snapshots.

cargo run -p tau-coding-agent -- \
  --model openai/gpt-5.2 \
  --gateway-contract-runner \
  --gateway-fixture crates/tau-gateway/testdata/gateway-contract/rollout-pass.json \
  --gateway-state-dir .tau/gateway \
  --gateway-guardrail-failure-streak-threshold 2 \
  --gateway-guardrail-retryable-failures-threshold 2

The runner writes state and observability output under:

.tau/gateway/state.json
.tau/gateway/runtime-events.jsonl
.tau/gateway/channel-store/gateway/<actor_id>/...

Inspect gateway transport health snapshot:

cargo run -p tau-coding-agent -- \
  --gateway-state-dir .tau/gateway \
  --transport-health-inspect gateway \
  --transport-health-json

Inspect gateway rollout guardrail/status report:

cargo run -p tau-coding-agent -- \
  --gateway-state-dir .tau/gateway \
  --gateway-status-inspect \
  --gateway-status-json

Operational rollout and rollback guidance: docs/guides/gateway-ops.md.

Deployment contract runner (cloud + WASM)

Use this fixture-driven runtime mode to validate cloud deployment and WASM rollout paths, retry outcomes, state persistence, and channel-store snapshots.

cargo run -p tau-coding-agent -- \
  --model openai/gpt-5.2 \
  --deployment-contract-runner \
  --deployment-fixture crates/tau-coding-agent/testdata/deployment-contract/rollout-pass.json \
  --deployment-state-dir .tau/deployment \
  --deployment-queue-limit 64 \
  --deployment-processed-case-cap 10000 \
  --deployment-retry-max-attempts 4 \
  --deployment-retry-base-delay-ms 0

The runner writes state and observability output under:

.tau/deployment/state.json
.tau/deployment/runtime-events.jsonl
.tau/deployment/channel-store/deployment/<blueprint_id>/...

Inspect deployment transport health snapshot:

cargo run -p tau-coding-agent -- \
  --deployment-state-dir .tau/deployment \
  --transport-health-inspect deployment \
  --transport-health-json

Inspect deployment rollout guardrail/status report:

cargo run -p tau-coding-agent -- \
  --deployment-state-dir .tau/deployment \
  --deployment-status-inspect \
  --deployment-status-json

Operational rollout and rollback guidance: docs/guides/deployment-ops.md.

No-code custom command contract runner (removed)

--custom-command-contract-runner has been removed from active transport dispatch. Preserve existing custom-command state directories and use inspection commands for diagnostics.

Inspection commands read state and observability output under:

.tau/custom-command/state.json
.tau/custom-command/runtime-events.jsonl
.tau/custom-command/channel-store/custom-command/<command_name or registry>/...

Inspect custom-command transport health snapshot:

cargo run -p tau-coding-agent -- \
  --custom-command-state-dir .tau/custom-command \
  --transport-health-inspect custom-command \
  --transport-health-json

Inspect custom-command rollout guardrail/status report:

cargo run -p tau-coding-agent -- \
  --custom-command-state-dir .tau/custom-command \
  --custom-command-status-inspect \
  --custom-command-status-json

Operational rollout and rollback guidance: docs/guides/custom-command-ops.md.

Voice interaction and wake-word contract runner

Use this fixture-driven runtime mode to validate voice wake-word detection, turn handling, retry outcomes, state persistence, and channel-store snapshots.

cargo run -p tau-coding-agent -- \
  --model openai/gpt-5.2 \
  --voice-contract-runner \
  --voice-fixture crates/tau-coding-agent/testdata/voice-contract/rollout-pass.json \
  --voice-state-dir .tau/voice \
  --voice-queue-limit 64 \
  --voice-processed-case-cap 10000 \
  --voice-retry-max-attempts 4 \
  --voice-retry-base-delay-ms 0

The runner writes state and observability output under:

.tau/voice/state.json
.tau/voice/runtime-events.jsonl
.tau/voice/channel-store/voice/<speaker_id>/...

Inspect voice transport health snapshot:

cargo run -p tau-coding-agent -- \
  --voice-state-dir .tau/voice \
  --transport-health-inspect voice \
  --transport-health-json

Inspect voice rollout guardrail/status report:

cargo run -p tau-coding-agent -- \
  --voice-state-dir .tau/voice \
  --voice-status-inspect \
  --voice-status-json

Operational rollout and rollback guidance: docs/guides/voice-ops.md.

Voice live session replay runner

Use this fixture-driven live mode to validate wake-word routing, live turn handling, and fallback behavior (invalid audio/provider outages).

cargo run -p tau-coding-agent -- \
  --model openai/gpt-5.2 \
  --voice-live-runner \
  --voice-live-input crates/tau-coding-agent/testdata/voice-live/single-turn.json \
  --voice-live-wake-word tau \
  --voice-live-max-turns 64 \
  --voice-live-tts-output \
  --voice-state-dir .tau/voice-live

The runner writes state and observability output under:

.tau/voice-live/state.json
.tau/voice-live/runtime-events.jsonl
.tau/voice-live/channel-store/voice/<speaker_id>/...

Inspect live voice transport health snapshot:

cargo run -p tau-coding-agent -- \
  --voice-state-dir .tau/voice-live \
  --transport-health-inspect voice \
  --transport-health-json

Inspect live voice rollout guardrail/status report:

cargo run -p tau-coding-agent -- \
  --voice-state-dir .tau/voice-live \
  --voice-status-inspect \
  --voice-status-json

Operational rollout and rollback guidance: docs/guides/voice-ops.md.

ChannelStore inspection and repair

Inspect one channel:

cargo run -p tau-coding-agent -- \
  --channel-store-root .tau/channel-store \
  --channel-store-inspect github/issue-9

Repair malformed JSONL lines for one channel:

cargo run -p tau-coding-agent -- \
  --channel-store-root .tau/channel-store \
  --channel-store-repair slack/C123

RPC protocol commands

Capabilities:

cargo run -p tau-coding-agent -- --rpc-capabilities

Validate one frame:

cargo run -p tau-coding-agent -- --rpc-validate-frame-file /tmp/rpc-frame.json

Dispatch one frame:

cargo run -p tau-coding-agent -- --rpc-dispatch-frame-file /tmp/rpc-frame.json

Dispatch NDJSON file:

cargo run -p tau-coding-agent -- --rpc-dispatch-ndjson-file /tmp/rpc-frames.ndjson

Serve long-lived NDJSON over stdin/stdout:

cat /tmp/rpc-frames.ndjson | cargo run -p tau-coding-agent -- --rpc-serve-ndjson

RPC schema compatibility fixtures live under crates/tau-coding-agent/testdata/rpc-schema-compat/.

Ownership

Primary ownership surfaces:

crates/tau-coding-agent (transport command dispatch and shared runtime entrypoints)
crates/tau-github-issues-runtime, crates/tau-slack-runtime, crates/tau-multi-channel (channel-specific runtimes)
crates/tau-gateway (gateway transport, auth, and remote access flows)
crates/tau-memory + crates/tau-agent-core (memory transport diagnostics and runtime ownership boundaries)

Ownership map: docs/guides/runbook-ownership-map.md.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Transport Guide

Runtime Log Rotation

Removed Contract Runner Migration Matrix

GitHub Issues bridge

Slack Socket Mode bridge

Multi-channel contract runner (Telegram, Discord, WhatsApp)

Multi-channel live runner (Telegram, Discord, WhatsApp)

Multi-agent contract runner

Semantic memory diagnostics (contract runner removed)

Browser automation live runner

Dashboard contract runner (removed)

Gateway OpenResponses + websocket control server

Tau daemon lifecycle

Gateway contract runner

Deployment contract runner (cloud + WASM)

No-code custom command contract runner (removed)

Voice interaction and wake-word contract runner

Voice live session replay runner

ChannelStore inspection and repair

RPC protocol commands

Ownership

FilesExpand file tree

transports.md

Latest commit

History

transports.md

File metadata and controls

Transport Guide

Runtime Log Rotation

Removed Contract Runner Migration Matrix

GitHub Issues bridge

Slack Socket Mode bridge

Multi-channel contract runner (Telegram, Discord, WhatsApp)

Multi-channel live runner (Telegram, Discord, WhatsApp)

Multi-agent contract runner

Semantic memory diagnostics (contract runner removed)

Browser automation live runner

Dashboard contract runner (removed)

Gateway OpenResponses + websocket control server

Tau daemon lifecycle

Gateway contract runner

Deployment contract runner (cloud + WASM)

No-code custom command contract runner (removed)

Voice interaction and wake-word contract runner

Voice live session replay runner

ChannelStore inspection and repair

RPC protocol commands

Ownership