CtrlK
BlogDocsLog inGet started
Tessl Logo

nitinjain999/platform-skills

Production-grade platform engineering handbook — Kubernetes, Terraform, Flux CD, GitHub Actions, AWS, and more.

67

Quality

84%

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

review.mdcommands/

name:
review
description:
Reviews a Kubernetes manifest, Terraform file, GitHub Actions workflow, Helm values, or any platform configuration for correctness, security, and operational safety. Supports bot comment mode for automated PR feedback.
argument-hint:
[paste file content or describe what to review] [--bot to emit GitHub-flavoured markdown for PR comments]

Interactive Wizard (fires when $ARGUMENTS is empty)

When invoked with no arguments, ask before reviewing:

Q1 — What should be reviewed?

Paste the file content to review, or describe what you want reviewed
(e.g. "my EKS Terraform module", "this GitHub Actions workflow", "Helm values file"):

Q2 — Output mode? (ask after Q1)

Output format:
  1. Standard   — narrative findings, for human review
  2. Bot / PR   — GitHub-flavoured markdown comment (use with --bot flag or in CI)

Enter 1 or 2 [default: 1]:

Then proceed with the review framework below using the provided content.


You are a senior platform engineer performing a production-readiness review.

Review the following: $ARGUMENTS

Evaluate in this priority order:

1. Correctness

  • Are API versions current and not deprecated?
  • Are required fields present?
  • Are references (namespaces, names, labels) consistent?
  • Will this actually do what the author intends?

2. Security

  • Is least-privilege applied? (RBAC, IAM, SCCs, network policy)
  • Are secrets handled safely? (no plaintext, no overly broad access)
  • Are containers running as non-root with read-only filesystems?
  • For GitHub Actions: are actions SHA-pinned, permissions minimal, no pull_request_target misuse?
  • For Terraform: are IAM policies scoped, no wildcard resources or actions?

3. Operational Safety

  • Is there a rollback path?
  • What is the blast radius if this fails?
  • Are resource limits and requests set?
  • Are health checks (liveness/readiness) defined?
  • For GitOps: is prune enabled? What happens on deletion?

4. Deprecations and Upgrade Risk

  • Any deprecated APIs, fields, or action versions?
  • Will this break on the next minor version of the tool?

5. Summary

Separate findings into:

  • Critical — must fix before merging
  • Improvement — should fix, not blocking
  • Note — informational only

Bot Comment Mode

When invoked with --bot or from a GitHub Actions context, emit the review as a GitHub-flavoured markdown PR comment using this exact structure so the workflow can post it with gh pr comment:

## 🔍 Platform Skills Review

<!-- platform-skills-review -->

### Result: {MERGE_READY | NEEDS_FIX | BLOCKED}

| Severity | Finding | File / Line |
|---|---|---|
| 🔴 Critical | <finding> | <file:line or n/a> |
| 🟡 Improvement | <finding> | <file:line or n/a> |
| 🔵 Note | <finding> | <file:line or n/a> |

#### Critical issues
<!-- one subsection per Critical finding with: problem, evidence, fix -->

#### Improvements
<!-- one subsection per Improvement with: problem, suggested fix -->

---
*Generated by [platform-skills](https://github.com/nitinjain999/platform-skills) · [docs](https://github.com/nitinjain999/platform-skills/wiki)*

Result values:

  • BLOCKED — one or more Critical findings
  • NEEDS_FIX — no Critical, but one or more Improvements
  • MERGE_READY — no findings or Notes only

The HTML comment <!-- platform-skills-review --> is used as a marker so the workflow can find and update the existing comment on subsequent pushes rather than posting a new one each time.

BEFORE_AFTER.md

CHANGELOG.md

CODE_OF_CONDUCT.md

COMMANDS.md

CONTRIBUTING.md

EDITOR_INTEGRATIONS.md

GETTING_STARTED.md

HOW_IT_WORKS.md

install.sh

INSTALLATION.md

LAUNCH.md

PROMPTS.md

QUICKSTART.md

README.md

renovate.json

SECURITY.md

SKILL.md

tessl.json

tile.json