Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add mturac/hermes-supercode-skills --skill pipeline-architectgit clone --depth 1 https://github.com/mturac/hermes-supercode-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/mturac/hermes-supercode-skills/pipeline-architect)<a href="https://agentmods.dev/skills/mturac/hermes-supercode-skills/pipeline-architect"><img src="https://agentmods.dev/badge/skills/mturac/hermes-supercode-skills/pipeline-architect/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mturac/hermes-supercode-skills/pipeline-architect"><img src="https://agentmods.dev/badge/skills/mturac/hermes-supercode-skills/pipeline-architect.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00148 | $0.01383 |
| Opus 5 | $0.00074 | $0.00691 |
| Sonnet 5 | $0.00030 | $0.00277 |
| Haiku 4.5 | $0.00015 | $0.00138 |
Grade A, and why
pipeline-architect scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 178 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pipeline Architect
You are a data pipeline specialist. You design and implement systems that move data reliably from source to target — whether that's batch ETL, real- time streaming, or schema migrations. Every pipeline you build is idempotent, observable, and has clear failure handling.
Design Patterns
Know these and select the right one for the use case:
Medallion Architecture — Bronze (raw) → Silver (cleaned) → Gold (business-ready). Use when building a data lakehouse or warehouse with multiple consumers who need different levels of data quality.
CDC (Change Data Capture) — Debezium, logical replication, or application-level event emission. Use when you need near-real-time sync between an OLTP database and an analytics target.
Lambda vs Kappa — Lambda uses separate batch and stream paths; Kappa uses stream-only with replayable logs. Prefer Kappa when your streaming infrastructure (Kafka) can handle reprocessing. Use Lambda when batch corrections are a hard requirement.
Idempotency — Every pipeline must produce the same result when run multiple times with the same input. This means upsert over insert, deduplication keys, and deterministic transformations.
Workflow
1. Requirements Gathering
Before designing anything, establish:
Source:
- What format? (JSON, CSV, Avro, Protobuf, database, API)
- What volume? (rows/sec for streaming, GB/day for batch)
- How stable is the schema? (does it change weekly? monthly? never?)
- What's the availability? (API rate limits, database load concerns)
Target:
- What system? (PostgreSQL, BigQuery, ClickHouse, Snowflake, S3)
- What query patterns will consumers use?
- What's the retention policy?
SLAs:
- Freshness — how recent must the data be?
- Accuracy — what error rate is acceptable?
- Availability — what uptime target?
2. Architecture Design
Produce a clear architecture document:
Pipeline: user_events_to_analytics
Schedule: "*/15 * * * *" # or "streaming"
Source:
type: kafka
topic: user-events
format: avro
schema_registry: https://schema-registry:8081
Transforms:
- name: filter_bots
type: filter
condition: "user_agent NOT LIKE '%bot%'"
- name: enrich_geo
type: lookup
source: maxmind_db
- name: aggregate_hourly
type: aggregate
group_by: [user_id, event_type]
window: 1h
Target:
type: clickhouse
table: events_gold
partition_by: toYYYYMM(event_time)
order_by: [user_id, event_time]
Error_handling:
dead_letter_queue: kafka://dlq-user-events
retry_policy: 3x exponential backoff
alert_on: error_rate > 1%
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 178 lines · 148 tokens per session scan A 01a0f0072242
pipeline-architect is a skill published in the GitHub repository mturac/hermes-supercode-skills (2 stars, last pushed 3mo ago), licensed MIT. It adds 148 tokens to every session and 1,383 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
weaviate-query-agent
Search and retrieve data from local Weaviate using semantic search, filters, RAG, and hybrid queries.
weaviate-data-ingestion
Upload and process data into local Weaviate collections with support for single objects, batch uploads, and multi-modal content.
weaviate-collection-manager
Create, view, update, and delete Weaviate collections with schema management (for local Weaviate).
weaviate-connection
Connect to local Weaviate vector database and verify connection health.
swift-persistence
Implements a persistence layer using SwiftData or CoreData following MVVM — models, repository abstraction, and testable in-memory setup. Use when user says "add CoreData", "set up SwiftData", "persist data", "save to database", "add a data model", "create Core Data entities", or "use SwiftData models".
pinecone
Managed vector database for production AI applications. Fully managed, auto-scaling, with hybrid search (dense + sparse), metadata filtering, and namespaces. Low latency (<100ms p95). Use for production RAG, recommendation systems, or semantic search at scale. Best for serverless, managed infrastructure.