All skills
nielsmadan avatar

/review-plan

@6661de7

Multi-agent review of implementation plans. Use after creating a plan but before implementing, especially for complex or risky changes.

Use this Skill: https://skilld.dev/gh/nielsmadan/agentic-coding/review-plan

This session only. Nothing lands on disk.

referencesrobustness-patterns.md

≈367 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Robustness Anti-Pattern Examples

Code examples of common fragile patterns and their robust alternatives. Used by the robustness review agent.

Timing Hacks (Never Do This)

// BAD: Timeout to "fix" race condition
await saveData();
await new Promise(r => setTimeout(r, 100)); // Hope it's done!
const data = await loadData();

// GOOD: Proper synchronization
await saveData();
await waitForSync(); // Explicit sync point
const data = await loadData();
// BAD: Polling for state change
let ready = false;
while (!ready) {
  await sleep(100);
  ready = checkIfReady();
}

// GOOD: Event-based
await new Promise(resolve => {
  emitter.once('ready', resolve);
});

Silent Failures

# BAD: Swallowing errors
try:
    result = risky_operation()
except:
    pass  # Hope it worked!

# GOOD: Handle or propagate
try:
    result = risky_operation()
except SpecificError as e:
    logger.error(f"Operation failed: {e}")
    raise OperationFailedError(e) from e

State Assumptions

// BAD: Assuming state is current
const user = cache.get(userId);
user.balance += amount;  // Stale!

// GOOD: Optimistic locking or transactions
const user = await db.transaction(async (tx) => {
  const u = await tx.users.findUnique({ where: { id: userId } });
  return tx.users.update({
    where: { id: userId, version: u.version },
    data: { balance: u.balance + amount, version: { increment: 1 } }
  });
});

Source: SKILL.md on GitHub

1 warning6mo4 checks · Risk SAFE
  • Gen Agent Trust Hub6mo

    The skill is safe to use and provides a structured framework for reviewing implementation plans using specialized sub-agents. It contains a potential surface for indirect prompt injection because it processes implementation plans and incorporates their content into sub-agent prompts without using boundary markers or sanitization.

  • Socket6mo

    No alerts

  • Snyk6mo

    Risk: MEDIUM · 1 issue

  • Runlayer7mo

    3 files scanned · No issues

Signed by skilld at 6661de7. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 3 weeks ago
argument-hint
[path to plan file or use current plan context]
effort
xhigh

README badge

README badge for nielsmadan/agentic-coding/review-plan