agent-tarballs.yml

name: Agent Tarballs

on:
  schedule:
    # 5 AM UTC daily
    - cron: "0 5 * * *"
  workflow_dispatch:
    inputs:
      agent:
        description: "Single agent to build (leave empty for all)"
        required: false
        type: string

permissions:
  contents: write

jobs:
  matrix:
    runs-on: ubuntu-latest
    outputs:
      agents: ${{ steps.set-matrix.outputs.agents }}
    steps:
      - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5  # v4

      - id: set-matrix
        env:
          AGENT_INPUT: ${{ inputs.agent }}
        run: |
          if [ -n "${AGENT_INPUT:-}" ]; then
            # Validate: agent name must be alphanumeric/hyphens only
            if ! printf '%s' "${AGENT_INPUT}" | grep -qE '^[a-z][a-z0-9-]*$'; then
              echo "::error::Invalid agent name: ${AGENT_INPUT}"
              exit 1
            fi
            echo "agents=$(jq -cn --arg a "${AGENT_INPUT}" '[$a]')" >> "$GITHUB_OUTPUT"
          else
            echo "agents=$(jq -c 'keys' packer/agents.json)" >> "$GITHUB_OUTPUT"
          fi

  build:
    needs: matrix
    runs-on: ${{ matrix.arch == 'arm64' && 'ubuntu-24.04-arm' || 'ubuntu-latest' }}
    strategy:
      max-parallel: 4
      fail-fast: false
      matrix:
        agent: ${{ fromJson(needs.matrix.outputs.agents) }}
        arch: [x86_64]
        # Native-binary agents need ARM builds too.
        # npm-based agents (codex, openclaw, kilocode) are arch-independent — x86_64 only.
        include:
          - agent: opencode
            arch: arm64
          - agent: hermes
            arch: arm64
          - agent: claude
            arch: arm64
          - agent: cursor
            arch: arm64
    steps:
      - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5  # v4

      - name: Install Bun
        uses: oven-sh/setup-bun@0c5077e51419868618aeaa5fe8019c62421857d6  # v2
        with:
          bun-version: "1.3.11"

      - name: Install agent under /root
        env:
          AGENT_NAME: ${{ matrix.agent }}
        run: |
          set -eo pipefail

          # Validate agent exists in agents.json (prevents path traversal / injection)
          if ! jq -e --arg a "${AGENT_NAME}" 'has($a)' packer/agents.json > /dev/null; then
            echo "::error::Unknown agent: ${AGENT_NAME}"
            exit 1
          fi

          # Read the agent's tier from packer/agents.json
          TIER=$(jq -r --arg a "${AGENT_NAME}" '.[$a].tier' packer/agents.json)
          echo "==> Agent: ${AGENT_NAME}, Tier: ${TIER}"

          # Run tier script (sets up node/bun/etc. under /root)
          if [ -f "packer/scripts/tier-${TIER}.sh" ]; then
            echo "==> Running tier script: tier-${TIER}.sh"
            sudo HOME=/root bash "packer/scripts/tier-${TIER}.sh"
          fi

          # TRUST BOUNDARY: packer/agents.json is version-controlled and requires
          # PR review to modify. Install commands are executed via bash -c, so any
          # change to agents.json MUST be reviewed carefully for command safety.
          #
          # Security layers:
          #   1. agents.json changes require PR review (branch protection)
          #   2. curl/wget targets validated against domain allowlist
          #   3. Suspicious command patterns rejected via blocklist
          #   4. Runs in ephemeral GitHub Actions container (destroyed after each run)
          echo "==> Installing agent..."

          # Allowed domains for curl/wget downloads (official agent vendor domains)
          ALLOWED_DOMAINS="claude.ai|cursor.com|opencode.ai|raw.githubusercontent.com|registry.npmjs.org|crates.io|github.com|dl.google.com"

          CMD_COUNT=$(jq -r --arg a "${AGENT_NAME}" '.[$a].install | length' packer/agents.json)
          i=0
          while [ "$i" -lt "$CMD_COUNT" ]; do
            cmd=$(jq -r --arg a "${AGENT_NAME}" --argjson i "$i" '.[$a].install[$i]' packer/agents.json)

            # Safety layer 1: reject suspicious command patterns
            if printf '%s' "${cmd}" | grep -qE '(mktemp|eval |base64 -d|/dev/tcp|nc -[elp]|python[23]? -c|perl -e|ruby -e|\bdd\b|>[[:space:]]*/dev/|rm -rf)'; then
              echo "::error::Suspicious install command rejected: ${cmd}"
              exit 1
            fi

            # Safety layer 2: validate curl/wget URLs against domain allowlist
            if printf '%s' "${cmd}" | grep -qE '(curl|wget)'; then
              urls=$(printf '%s' "${cmd}" | grep -oE 'https?://[^[:space:]"|'\'']+' || true)
              for url in ${urls}; do
                domain=$(printf '%s' "${url}" | sed -E 's|^https?://([^/]+).*|\1|')
                if ! printf '%s' "${domain}" | grep -qE "^(${ALLOWED_DOMAINS})$"; then
                  echo "::error::curl/wget to unapproved domain '${domain}' in: ${cmd}"
                  exit 1
                fi
              done
            fi

            echo "==> Running: ${cmd}"
            sudo HOME=/root bash -c "${cmd}" < /dev/null
            i=$((i + 1))
          done

      - name: Capture agent files into tarball
        env:
          AGENT_NAME: ${{ matrix.agent }}
        run: |
          set -eo pipefail
          sudo bash packer/scripts/capture-agent.sh "${AGENT_NAME}"

      - name: Create or update GitHub Release
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          AGENT_NAME: ${{ matrix.agent }}
        run: |
          set -eo pipefail
          TAG="agent-${AGENT_NAME}-latest"
          DATE=$(date -u +%Y%m%d)
          ARCH="${{ matrix.arch }}"
          TARBALL="spawn-agent-${AGENT_NAME}-${ARCH}-${DATE}.tar.gz"

          # Move tarball to expected name (tarball is owned by root from sudo capture)
          sudo mv "/tmp/spawn-agent-${AGENT_NAME}.tar.gz" "${TARBALL}"
          sudo chown "$(id -u):$(id -g)" "${TARBALL}"

          # Create release if it doesn't exist, then upload the arch-specific tarball.
          # Multiple arch builds (x86_64, arm64) upload to the same release.
          if ! gh release view "${TAG}" > /dev/null 2>&1; then
            gh release create "${TAG}" \
              --title "Agent tarball: ${AGENT_NAME} (${DATE})" \
              --notes "Pre-built tarball for \`${AGENT_NAME}\` agent. Auto-generated nightly." \
              --prerelease
          fi

          # Delete stale asset for this arch if present (from a previous build today)
          gh release delete-asset "${TAG}" "${TARBALL}" --yes 2>/dev/null || true
          # Also clean up any older-dated tarball for this arch
          # grep returns exit 1 when no matches — pipe through cat to avoid pipefail killing the step
          gh release view "${TAG}" --json assets --jq ".assets[].name" 2>/dev/null \
            | { grep "spawn-agent-${AGENT_NAME}-${ARCH}-" || true; } \
            | while IFS= read -r old; do
                gh release delete-asset "${TAG}" "${old}" --yes 2>/dev/null || true
              done

          gh release upload "${TAG}" "${TARBALL}"