Slice 241 completed and pushed.

Commit: 71e8893

Validation evidence:

python3 -m unittest -v apps/agent-runtime/tests/test_runtime_invariants.py -> Ran 5 tests, OK
python3 -m unittest -v apps/agent-runtime/tests/test_runtime_services.py -> Ran 47 tests, OK
python3 -m unittest -v apps/agent-runtime/tests/test_runtime_adapters.py -> Ran 13 tests, OK
python3 -m unittest -v apps/agent-runtime/tests/test_trade_path.py -> Ran 151 tests, OK
python3 -m unittest -v apps/agent-runtime/tests/test_liquidity_cli.py -> Ran 21 tests, OK
python3 -m unittest -v apps/agent-runtime/tests/test_x402_cli.py -> Ran 4 tests, OK
final combined regression sweep: python3 -m unittest -v apps/agent-runtime/tests/test_trade_path.py apps/agent-runtime/tests/test_liquidity_cli.py apps/agent-runtime/tests/test_x402_cli.py apps/agent-runtime/tests/test_runtime_invariants.py -> Ran 181 tests, OK
npm run db:parity -> checkedAt=2026-03-08T21:07:00.617Z
npm run seed:reset -> ok: true
npm run seed:load -> ok: true, loadedAt=2026-03-08T21:07:06.453Z
npm run seed:verify -> ok: true
restored tracked infrastructure/seed-data/.seed-state.json after seed commands
npm run build -> success
pm2 restart all -> xclaw-web online

Code evidence:

apps/agent-runtime/xclaw_agent/commands/trade.py: cmd_trade_execute(...) now enforces mode=real before either EVM or Solana execution, closing the Solana mock-execution gap.
apps/agent-runtime/tests/test_trade_path.py: added command-surface failure injection for trade status posting, native transfer approval sync failure, Solana mock rejection, and approvals x402 fallback mirror failure.
apps/agent-runtime/tests/test_liquidity_cli.py: added liquidity status-post failure coverage.
apps/agent-runtime/tests/test_x402_cli.py: added best-effort outbound mirror failure coverage.
apps/agent-runtime/tests/test_runtime_invariants.py: added EVM/Solana mock-mode rejection invariant coverage.

Slice 241: Command-Surface Failure Injection Sweep #94

Description