[muhammad-tahir-nawaz]

The seed step writes nats-nkey.seed at the job root but exports NATS_NKEY_SEED_FILE=./nats-nkey.seed. The server and the test runner script run in inputs.working_directory, so for a monorepo service (e.g. asset-pipeline services/asset-normalizer) the relative path points at a file that does not exist: failed to load NKey seed from ./nats-nkey.seed.

What

Export $PWD/nats-nkey.seed. The file stays where it was, so callers using the default working directory get the same file, and existing overrides (../../nats-nkey.seed in asset-pipeline, ./nats-nkey.seed in pulse, blank in fleet-manager/worker-systems) are unaffected.

[Copilot]

🟢 Approval recommended

The focused change correctly preserves seed creation while making it accessible from alternate working directories.

0 open findings

What changed in this PR

Exports the generated NATS seed file using an absolute path so integration tests work from nested service directories.

Changes:

• Documents why an absolute path is required.
• Writes the absolute seed path to GITHUB_ENV.

File	Description
.github/​workflows/​integration-tests.yaml	Exports the root-level NATS seed file with an absolute path.

🧠 Review effort: Balanced

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[greptile-apps]

Confidence Score: 5/5

[Medium impact] This change appears safe to merge.

Summary

Exports NATS_NKEY_SEED_FILE as an absolute path so servers and tests can find the seed from a service subdirectory.

• NATS callers can find the seed file from a nested service folder.

_{Reviews (1) · Last reviewed commit: "fix: export the nats seed file as an abs..." · Reviewed by Greptile}