l3_live_context.sh

#!/usr/bin/env bash
# shellcheck source-path=SCRIPTDIR
# L3 - Agent + live context.
#
# Upgrades your agent with read-only SQL tools from an MCP server and opens a
# conversation. Ask, run the injector in a second terminal, then ask again: the
# answer changes.
#
#   ./l3_live_context.sh
#
# Needs ork and jq. The injector is `python inject.py` or `npm run inject`.
set -euo pipefail
# shellcheck source=lib.sh
. "$(dirname "$0")/lib.sh"
hello_setup ORCA_BASE_URL ORCA_MODEL

QUESTION="Which accounts look like an account takeover right now?"

ensure_environment "hello-env-$HELLO_PARTICIPANT"

# The same agent, next version: agent/<stack>/l3-live-context.json adds the MCP server.
#      ork agent update <id> --version <v> --mcp-server name=<name>,type=url,url=<MCP URL> --tool-json <toolset>
ensure_agent l3-live-context
echo "$AGENT_NAME v$AGENT_VERSION: $(jq -r .summary "$(layer_file l3-live-context)")"

# StreamNative Cloud's MCP server needs a credential. It goes in a vault, never
# in the prompt. The MCP server on your laptop takes none, so there is no vault.
#      ork agent vaults create --display-name hello-vault-<you>
#      ork agent vaults credentials create --vault <id> --mcp-server-url <SN_MCP_URL>
mcp_vault

#      ork agent sessions create ... [--vault-id <id>]
create_session "L3: live context" "$VAULT_ID"
# shellcheck disable=SC2016  # the backticks are for the reader
printf 'Tip: after the first answer, run `python inject.py` (or `npm run inject`) in another terminal and ask again.\n\n'
chat "$SESSION_ID" "$QUESTION"