jmcqueen/collabSupport
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
collabSupport/.env.example
jmcqueen 44ab6559ac Default WAN window to 24h + surface /voicediag in /help 
Bumps the shared default WAN look-back from 15m to 24h (1440m) so
both /phonestatus follow-ups and /voicediag pick up a full day of
voice-quality signal by default — better for after-the-fact ticket
triage than a live snapshot. Operators wanting real-time behavior
can set WAN_STANDARD_WINDOW_MINUTES=15 or pass --window 15m to
/voicediag.

Also fills in a long-standing help gap: /voicediag was fully
implemented but never listed in /help. Adds it under "AV & phones"
with usage, examples, and notes covering detail mode, --only,
--window, thresholds, kill switch, HTTP shape, and audit logging.
Updates /help phonestatus to call out the 24h WAN follow-up +
cross-link to /voicediag.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-07-09 09:50:30 -04:00

384 lines
19 KiB
Text
Raw Blame History

# =============================================================================
# CollabFinder / CollabSupport Environment Variables
# Copy this file to .env and fill in your values.
# =============================================================================
# -----------------------------------------------------------------------------
# Server
# -----------------------------------------------------------------------------
SERVER_PORT=1800
# Logging level: info (default - clean), debug (verbose, includes per-fetch details)
LOG_LEVEL=info
# Verbose Webex framework debug logs. Default off; auto-enabled when LOG_LEVEL=debug.
# WEBEX_FRAMEWORK_DEBUG=false
# -----------------------------------------------------------------------------
# HTTP API authentication
# -----------------------------------------------------------------------------
# Shared secret required to call destructive /:command HTTP endpoints such as
# /offboarduser, /provision-dect, /provision-vc, /vcmonitor, /bulkavstatuscsv,
# /bulkavswitchcsv, /devicesbymodel. Without it, those endpoints fail-closed
# with HTTP 503 — set this to any high-entropy string (e.g. `openssl rand -hex 32`).
# Callers send the token as `Authorization: Bearer <token>` or `X-API-Token: <token>`.
HTTP_API_TOKEN=
# Optional. When set to "true" / "1" / "yes", the same token is also required
# for read-only endpoints (/avstatus, /phonestatus, /av/devices/build/…,
# /phone/devices/build/…, /api/av/*, etc.). Default = false (those endpoints
# stay open so the bundled dashboards keep working without auth headers).
HTTP_API_REQUIRE_AUTH=false
# -----------------------------------------------------------------------------
# Webex Bot (required)
# -----------------------------------------------------------------------------
# Bot token from developer.webex.com. The framework connects to Webex over
# websockets using this token, so no public ingress / webhook URL is needed.
WEBEX_BOT_TOKEN=your-bot-token-here
# Service App credentials (used by WebexServiceAppAuth for backend Webex API
# calls — people lookup, room operations, etc.). Separate from the bot token.
#
# Required scopes (set when creating the service app at developer.webex.com):
# - spark-admin:people_read (people lookup, voicediag reads,
# webexhost, findEmptyLocations)
# - spark-admin:people_write (webexhost: assign licenses,
# voicediag: apply remediations —
# disable DND / clear forwarding /
# disable intercept / enable call
# waiting via /v1/people/{id}/features/*)
# - spark-admin:licenses_read (webexhost, findEmptyLocations)
# - identity:tokens_read (offboarduser: list a user's authorizations)
# - identity:tokens_write (offboarduser: revoke a user's authorizations)
# The authorizing admin must also hold Full / User / Device Admin role for the
# token-management calls to succeed.
#
# NOTE — early drafts of /voicediag documented `spark-admin:telephony_config_*`
# scopes and a `/v1/telephony/config/people/{id}/callSettings/*` URL scheme.
# That URL family returns 404 "no static resource" from the Webex gateway.
# The live admin surface is `/v1/people/{id}/features/{feature}` (see
# services/voiceDiag/ for details) and it uses the same `spark-admin:people_*`
# scopes that /webexhost and /offboarduser already require — no additional
# scope work is needed to enable /voicediag.
#
# Refresh tokens preserve their original scope set, so if you ever DO add a
# new scope you must:
# 1. Save the updated scopes on the service app.
# 2. Re-authorize the app for the org as a Full / User admin.
# 3. Delete/rotate tokens/webex-service-tokens.json so the next request
# re-bootstraps with the enlarged scope set.
WEBEX_CLIENT_ID=your-service-app-client-id
WEBEX_CLIENT_SECRET=your-service-app-client-secret
# Path to the rotating service app tokens file (must be writable).
# IMPORTANT: Use a *relative* path (e.g. ./config/...). The same .env works for both:
# - Local runs (resolved against your project root cwd)
# - Docker (resolved against /app inside container; see docker-compose volume mount)
# Do NOT use an absolute host path here — it will break inside the container.
WEBEX_TOKENS_PATH=./config/webex-service-tokens.json
# Optional override for the Webex API base URL (default https://webexapis.com/v1).
# WEBEX_BASE_URL=https://webexapis.com/v1
# -----------------------------------------------------------------------------
# Palo Alto Prisma SD-WAN (formerly CloudGenix) — WAN metrics for /voicediag
# and /phonestatus follow-up
# -----------------------------------------------------------------------------
# The Prisma integration is optional: leave it unconfigured and the WAN
# check bucket in /voicediag skips cleanly ("not applicable — missing
# sdwanSite") and /phonestatus omits the WAN follow-up message. Enable it
# by setting PRISMA_AUTH_MODE + the matching credential block below.
#
# AE convention (encoded in integrations/paloalto/sites.js): Prisma site
# names are `CG${storeNum.padStart(5, '0')}` — store 782 → CG00782, store
# 2477 → CG02477. Exact match. No fuzzy searching.
#
# --- Auth mode selection ---
# PRISMA_AUTH_MODE=sase (default, recommended for production)
# PRISMA_AUTH_MODE=legacy (only if you inherited a CloudGenix service
# account and can't provision a SASE one)
# PRISMA_AUTH_MODE=
# --- Unified SASE OAuth 2.0 (PRISMA_AUTH_MODE=sase) ---
# Provision a service-app credential in Strata Cloud Manager → Identity &
# Access → Service Accounts. TSG_ID is the Tenant Service Group id shown
# in the same panel — it scopes the resulting bearer token to your tenant.
# Both defaults below are correct for the public SASE cloud; override for
# region-specific gateways if you're on one.
# PRISMA_CLIENT_ID=
# PRISMA_CLIENT_SECRET=
# PRISMA_TSG_ID=
# PRISMA_SASE_BASE_URL=https://api.sase.paloaltonetworks.com
# PRISMA_AUTH_URL=https://auth.apps.paloaltonetworks.com/oauth2/access_token
# --- Legacy CloudGenix session token (PRISMA_AUTH_MODE=legacy) ---
# Interactive-user credential (email + password). Kept for backward
# compatibility only — the SASE mode above is what new deployments should
# use because it's service-account-shaped instead of impersonating a
# human. Session tokens last ~8 hours; the client transparently
# re-authenticates on 401.
# PRISMA_EMAIL=
# PRISMA_PASSWORD=
# PRISMA_LEGACY_BASE_URL=https://api.cloudgenix.com
# --- WAN voice-quality thresholds (used by /voicediag WAN bucket + the
# /phonestatus WAN follow-up renderer) ---
# Defaults follow ITU-T G.114 (latency), G.711 PLC (loss), and RFC 3550
# (jitter) references. See services/voiceDiag/README.md for the standards
# reference table.
# WAN_STANDARD_LATENCY_WARN_MS=150
# WAN_STANDARD_LATENCY_ERROR_MS=400
# WAN_STANDARD_JITTER_WARN_MS=30
# WAN_STANDARD_JITTER_ERROR_MS=50
# WAN_STANDARD_LOSS_WARN_PCT=1
# WAN_STANDARD_LOSS_ERROR_PCT=3
# WAN_STANDARD_MOS_WARN=4.0
# WAN_STANDARD_MOS_ERROR=3.5
# WAN_STANDARD_HEALTHSCORE_WARN=80
# WAN_STANDARD_HEALTHSCORE_ERROR=60
# --- WAN bucket global kill-switch (mirror of VOICE_STANDARD_ENABLED for
# the port bucket). Set to `false` to silence the entire WAN check
# bucket while a Prisma cleanup is in progress. The /phonestatus
# follow-up is unaffected — it always renders whatever Prisma
# returns. Feature-config + port-hygiene buckets keep running. ---
# WAN_STANDARD_ENABLED=true
# --- WAN look-back window (minutes) used when the operator doesn't
# pass `--window` on /voicediag (and always used by /phonestatus).
# Applies to healthscore + LQM. Alarms window is
# max(60, WAN_STANDARD_WINDOW_MINUTES) since sub-hour alarm queries
# are usually too noisy to be actionable. Prisma's finest LQM
# bucket is 5m; anything <= 5 is snapped up.
# Common values: 15 (real-time), 60, 360 (6h), 1440 (24h — default).
# WAN_STANDARD_WINDOW_MINUTES=1440
# -----------------------------------------------------------------------------
# /voicediag — Store voice-line standards
# -----------------------------------------------------------------------------
# Store phones are held to a fixed configuration standard. Most of the
# standard is baked into the per-check modules under
# services/voiceDiag/checks/ (see services/voiceDiag/README.md for the
# reference table) — the values below only exist for the pieces that
# vary per site / per environment.
#
# Expected VLAN for store phones. Today AE tags phones into the data
# VLAN (102); some sites may move onto a proper voice VLAN in the
# future. The check compares against whatever value is set here and
# emits a warn when a phone shows up on any other VLAN. Default 102
# if unset.
# VOICE_STANDARD_PHONE_VLAN=102
# Global kill-switch for /voicediag port-hygiene checks. Set to
# `false` to silence portType / portVlan / portPoe / portEnabled while
# an underlying Meraki cleanup is in progress and you don't want the
# noise. Values: true (default) / false. The feature-config checks
# (DND, forwarding, waiting, voicemail, intercept, hoteling, exec-
# assistant, outgoingPermission) always run — this only gates the
# switchport bucket.
# VOICE_STANDARD_ENABLED=true
# -----------------------------------------------------------------------------
# /webexhost — Webex Meetings host license helper
# -----------------------------------------------------------------------------
# Site to evaluate host status against. Default: aeo2go.webex.com.
# WEBEX_HOST_SITE_URL=aeo2go.webex.com
# License ID auto-assigned by `/webexhost <email>` when the user is missing a
# host license on the site. Discover the right id by running `/webexhost list`
# (lists every meeting license on the site with id + remaining seats).
# Until this is set, `/webexhost <email>` will still report status, but the
# confirm-assign step refuses with a clear message pointing at /webexhost list.
WEBEX_HOST_LICENSE_ID=
# -----------------------------------------------------------------------------
# Jira (required for /jira* commands)
# Use JIRA_CLOUD_ID for service accounts / new Atlassian API gateway endpoints:
# JIRA_CLOUD_ID=<uuid-from-atlassian>
# (constructs https://api.atlassian.com/ex/jira/<id>/rest/api/3 ...)
# Otherwise fall back to classic site base:
# JIRA_BASE_URL=https://your-org.atlassian.net
# -----------------------------------------------------------------------------
JIRA_CLOUD_ID=
JIRA_BASE_URL=https://your-org.atlassian.net
JIRA_EMAIL=your-email@company.com
JIRA_API_TOKEN=your-jira-api-token
JIRA_MAX_RESULTS=30
# -----------------------------------------------------------------------------
# Jira Poller (hourly ticket enrichment)
# -----------------------------------------------------------------------------
# Cron poller that scans unassigned tickets in the AV / Comm Services /
# Mobility queue every hour, posts a phone or AV status snapshot as a
# Jira comment on each store-scoped ticket, labels the ticket
# `bot-enriched` so it's not re-processed, and posts a summary of newly
# enriched tickets to a Webex space. Tickets the AI classifier decides
# are out-of-scope get labeled `bot-skipped` (same anti-retry purpose,
# no comment posted) so we don't burn AI tokens re-reading them every
# hour.
#
# Required scopes on JIRA_API_TOKEN: read + write on issues in the
# target projects (comment + edit-labels). The token owner needs "Add
# Comments" and "Edit Issues" permission — a plain read-only integration
# token WILL NOT work.
#
# JIRA_POLLER_ROOM_ID
# Webex space roomId to post the per-poll "N new tickets" summary to.
# Poller stays DISABLED (cron never registered) if unset — safe default
# for dev instances that share the same Jira credentials.
#
# JIRA_POLLER_PRIME_ON_START
# One-shot backlog-prime toggle. Set to `true` for a SINGLE deploy to
# bulk-label every ticket currently matching the poller's JQL as
# `bot-enriched` WITHOUT enriching them or posting a summary. Prevents
# day-one spam from a queue that already has dozens of open tickets.
# Flip back to `false` (or remove) before the next restart or the
# prime pass runs again.
#
# JIRA_STORE_FIELD_ID
# Optional. Numeric custom-field id for the `Store Number` field
# (e.g. `customfield_10042`). If unset, the poller discovers it at
# first use via GET /rest/api/3/field. Set explicitly to skip
# discovery (saves one API call at startup) or when the display
# name resolves ambiguously in your Jira schema.
#
# Note: on tenants where Store Number is an Atlassian Assets object
# reference (not a plain string), the field value the poller reads
# will be an opaque object like {"objectId":"81255"}. The AI
# classifier handles this by extracting the store number from the
# ticket summary / description text instead ("Store 3860 - ..."), so
# the field being unreadable is not fatal.
#
# JIRA_POLLER_MODEL
# Optional model override for the AI ticket classifier. Defaults to
# XAI_MODEL if unset. Classification is a small, deterministic
# structured task (~50-token JSON output per ticket) that doesn't
# need the reasoning depth of the summary model — a cheaper/faster
# model (e.g. `grok-3-mini`) saves noticeable money at scale without
# hurting classification accuracy on the phone/av/skip taxonomy.
# -----------------------------------------------------------------------------
JIRA_POLLER_ROOM_ID=
JIRA_POLLER_PRIME_ON_START=false
JIRA_STORE_FIELD_ID=
JIRA_POLLER_MODEL=
# -----------------------------------------------------------------------------
# xAI / Grok (used for ticket and work order summarization)
# -----------------------------------------------------------------------------
XAI_URL=https://api.x.ai/v1/chat/completions
XAI_API_KEY=your-xai-api-key
XAI_MODEL=grok-2-latest
# -----------------------------------------------------------------------------
# Meraki
# -----------------------------------------------------------------------------
MERAKI_API_KEY=your-meraki-api-key
MERAKI_ORG_ID=your-org-id
# -----------------------------------------------------------------------------
# RED (Digital Signage)
# Comma-separated list of company IDs (one per company tenant).
# -----------------------------------------------------------------------------
RED_BASE_URL=https://api.red.com
RED_CLIENT_ID=your-red-client-id
RED_API_KEY=your-red-api-key
RED_COMPANY_IDS=company-id-1,company-id-2,company-id-3
# -----------------------------------------------------------------------------
# OptiSigns
# -----------------------------------------------------------------------------
OPTISIGN_API_KEY=your-optisigns-api-key
# -----------------------------------------------------------------------------
# DigiCert (for VC provisioning)
# -----------------------------------------------------------------------------
DIGICERT_API_KEY=your-digicert-key
DIGICERT_BASE_URL=https://one.digicert.com
DIGICERT_PROFILE_ID=your-profile-id
DIGICERT_SEAT_EMAIL=your-email@company.com
# -----------------------------------------------------------------------------
# MDM / Workspace ONE (two instances)
# -----------------------------------------------------------------------------
# Standard / Store MDM
WS1_CLIENT_ID=...
WS1_CLIENT_SECRET=...
WS1_TENANT_CODE=...
# CORP MDM (used for offboarding / enterprise wipes)
CORP_WS1_API_BASE=https://...
CORP_WS1_CLIENT_ID=...
CORP_WS1_CLIENT_SECRET=...
CORP_WS1_TENANT_CODE=...
# -----------------------------------------------------------------------------
# Atlas
# -----------------------------------------------------------------------------
ATLAS_AUTH_KEY=your-atlas-key
# -----------------------------------------------------------------------------
# ServiceChannel
# OAuth password grant against ServiceChannel's identity endpoint.
# -----------------------------------------------------------------------------
SC_BASE_URL=https://api.servicechannel.com/v3
SC_OAUTH_URL=https://login.servicechannel.com/oauth/token
SC_CLIENT_ID=your-sc-client-id
SC_CLIENT_SECRET=your-sc-client-secret
SC_USERNAME=your-sc-username@company.com
SC_PASSWORD=your-sc-password
# -----------------------------------------------------------------------------
# VC Provisioning "backdoor" account
# Used by vcProvisionService for local-device authentication during certificate
# enrollment. DIGICERT_SEAT_EMAIL above is unrelated.
# -----------------------------------------------------------------------------
BACKDOOR_USERNAME=monitor
BACKDOOR_PASSWORD=...
# -----------------------------------------------------------------------------
# Cisco DBS-210 DECT — LOCAL DEV TEST HARNESS
# Used only by scripts/testDectBase.js when iterating locally against a
# lab base. NOT read by the bot at runtime — the bot never talks to a
# DBS-210 directly. Actual production DBS-210 access lives in the
# DECT_RELAY_* env below and is executed by dect-relay-agent/.
#
# Cisco's guidance: use the DECT serviceability password (Control Hub
# → Calling → Features → DECT Networks → Manage → Manage DECT
# serviceability password). Our tenant is configured to share a
# single password across all bases in the fleet.
# -----------------------------------------------------------------------------
DECT_TEST_BASE_IP=10.0.0.100
DECT_TEST_USER=admin
DECT_TEST_PASSWORD=your-dect-serviceability-password
# Optional: seconds to wait for base station responses. DBS-210 is slow
# on syslog/PRT downloads; 30s is a good starting point.
DECT_TEST_TIMEOUT_MS=30000
# -----------------------------------------------------------------------------
# DECT Relay Hub — WSS endpoint for the data-center relay agent
# The bot process runs in the public cloud and can't reach 10.x. The
# relay agent (see dect-relay-agent/) runs INSIDE the DC, dials
# outbound over WSS to this bot, and executes any DECT command the bot
# pushes. Feature-gated: without DECT_RELAY_AGENT_TOKEN, the WSS
# endpoint is not attached and /phonestatus's DECT follow-up + the
# future /dectstatus command return "relay not connected".
#
# Rotate DECT_RELAY_AGENT_TOKEN on both sides at once. Suggested
# generation: `openssl rand -hex 32`.
# -----------------------------------------------------------------------------
DECT_RELAY_AGENT_TOKEN=
# Optional. Default: /dect-relay/ws. Change only if you also change
# DECT_RELAY_BOT_URL on the agent side to match.
# DECT_RELAY_PATH=/dect-relay/ws
# Optional. Per-base collect() RPC timeout. Corporate proxies can make
# DBS-210 reads slow; 15s is comfortable, 30s is generous.
# DECT_COLLECT_TIMEOUT_MS=15000
# -----------------------------------------------------------------------------
# Notes
# -----------------------------------------------------------------------------
# - config/config.json has been fully removed. All configuration is via env vars.
# - The rotating Webex service token lives in config/webex-service-tokens.json
# and is mounted separately when running in Docker.
# - Add any new integration keys above following the same pattern.
Reference in a new issue View git blame Copy permalink