Finta Performance TuningSAFE

Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: finta-performance-tuning
description: 'Optimize Finta fundraise workflow efficiency.

  Trigger with phrases like "finta performance", "finta efficiency", "optimize finta".

  '
allowed-tools: Read, Grep
version: 1.7.0
license: MIT
author: Jeremy Longshore <[email protected]>
tags:
- saas
- fundraising-crm
- investor-management
- finta
compatibility: Designed for Claude Code
---
# Finta Performance Tuning

## Overview

Finta's fundraising API handles investor list pagination, round data aggregation, and CRM sync batching. Founders querying large investor databases (1,000+ contacts) hit pagination bottlenecks, while round aggregation across multiple funding stages compounds latency. Optimizing paginated fetches with cursor-based iteration, caching investor profiles, and batching CRM sync writes reduces pipeline load times by 50-70% and keeps fundraising dashboards responsive during active rounds.

## Prerequisites

- A baseline from aggregate, redacted latency and error metrics; do not copy investor records into performance traces.
- An approved capacity target and freshness expectation for each dashboard or sync.
- A staging environment with synthetic data and a rollback switch for cache, queue, and concurrency changes.

## Instructions

1. Measure one bottleneck at a time: pagination, cache misses, queue depth, or destination latency.
2. Set bounded concurrency and exponential backoff before increasing throughput; respect provider responses rather than assuming a rate limit.
3. Cache only data that is permitted to persist and define invalidation on the write path; do not trade data isolation for a higher hit rate.
4. Roll out a small canary, compare redacted metrics against the baseline, and revert if error rate, staleness, or queue age breaches the agreed threshold.
5. Record the observed result and owner so the tuning change can be reviewed or removed later.

## Caching Strategy

```typescript
const cache = new Map<string, { data: any; expiry: number }>();
const TTL = { investors: 600_000, rounds: 300_000, pipeline: 120_000 };

async function cached(key: string, ttlKey: keyof typeof TTL, fn: () => Promise<any>) {
  const entry = cache.get(key);
  if (entry && entry.expiry > Date.now()) return entry.data;
  const data = await fn();
  cache.set(key, { data, expiry: Date.now() + TTL[ttlKey] });
  return data;
}
// Investor profiles change rarely (10 min). Pipeline stages are volatile (2 min).
```

## Batch Operations

```typescript
async function syncInvestorsBatch(client: any, cursor?: string, pageSize = 100) {
  const allInvestors = [];
  let nextCursor = cursor;
  do {
    const page = await client.listInvestors({ cursor: nextCursor, limit: pageSize });
    allInvestors.push(...page.data);
    nextCursor = page.next_cursor;
    if (nextCursor) await new Promise(r => setTimeout(r, 200));
  } while (nextCursor);
  return allInvestors;
}
```

## Connection Pooling

```typescript
import { Agent } from 'https';
const agent = new Agent({ keepAlive: true, maxSockets: 8, maxFreeSockets: 4, timeout: 30_000 });
// Finta API calls are lightweight — moderate socket count suffices
```

## Rate Limit Management

```typescript
async function withRateLimit(fn: () => Promise<any>): Promise<any> {
  const res = await fn();
  const remaining = parseInt(res.headers?.['x-ratelimit-remaining'] || '50');
  if (remaining < 3) {
    const resetMs = parseInt(res.headers?.['x-ratelimit-reset'] || '5') * 1000;
    await new Promise(r => setTimeout(r, resetMs));
  }
  return res;
}
```

## Monitoring

```typescript
const metrics = { apiCalls: 0, cacheHits: 0, syncErrors: 0, avgLatencyMs: 0 };
function track(startMs: number, cached: boolean, error?: boolean) {
  metrics.apiCalls++;
  metrics.avgLatencyMs = (metrics.avgLatencyMs * (metrics.apiCalls - 1) + (Date.now() - startMs)) / metrics.apiCalls;
  if (cached) metrics.cacheHits++; if (error) metrics.syncErrors++;
}
```

## Performance Checklist

- [ ] Use cursor-based pagination for investor lists (not offset)
- [ ] Cache investor profiles with 10-min TTL
- [ ] Batch CRM sync writes in groups of 50
- [ ] Aggregate round data client-side to avoid repeated queries
- [ ] Enable HTTP keep-alive for persistent connections
- [ ] Parse rate limit headers and pause before exhaustion
- [ ] Prefetch deal room analytics during idle periods
- [ ] Set pipeline cache TTL to 2 min for active-round freshness

## Error Handling

| Issue | Cause | Fix |
|-------|-------|-----|
| Slow investor list load | Offset-based pagination on large dataset | Switch to cursor-based iteration with limit=100 |
| Stale round totals | Aggregation cache too long during active round | Reduce round TTL to 5 min, invalidate on write |
| CRM sync timeout | Too many individual writes | Batch CRM updates in groups of 50 |
| 429 Rate Limited | Burst of API calls during pipeline refresh | Parse rate limit headers, add progressive backoff |

## Output

Publish a performance receipt with the baseline and post-change aggregate metrics, cache and queue settings, test window, decision owner, and rollback status. Exclude contact details, document links, investment amounts, and raw payloads from the receipt.

## Examples

In staging, replay a synthetic page sequence at low concurrency, then raise it one step while watching queue age and error percentage. If the destination returns a throttle response, reduce concurrency and verify that the retry queue drains without duplicating a synthetic record before considering a production canary.

## Resources

- Finta Developer Docs
- [Finta Blog](https://www.trustfinta.com/blog)

## Next Steps

See `finta-reference-architecture`.

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

What the source does

Filesystem

none-observed

Network

none-observed

Shell

none-observed

Dependencies

pinned

Secrets in source

none-found

Findings (0)

No findings outside the package's declared scope.

Gates applied: no_behavioural_pass.