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 vaquarkhan/data-engineering-agent-skills --skill cdc-and-incremental-loadinggit clone --depth 1 https://github.com/vaquarkhan/data-engineering-agent-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/vaquarkhan/data-engineering-agent-skills/cdc-and-incremental-loading)<a href="https://agentmods.dev/skills/vaquarkhan/data-engineering-agent-skills/cdc-and-incremental-loading"><img src="https://agentmods.dev/badge/skills/vaquarkhan/data-engineering-agent-skills/cdc-and-incremental-loading.svg" alt="Measured on agentmods" 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 | $0.00048 | $0.00507 |
| Opus 5 | $0.00024 | $0.00253 |
| Sonnet 5 | $0.00010 | $0.00101 |
| Haiku 4.5 | $0.00005 | $0.00051 |
Grade A, and why
cdc-and-incremental-loading 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 8d 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CDC And Incremental Loading
Overview
Use this skill when the system changes data over time and full reloads are not the right answer. It helps agents design safe incremental behavior with clear watermarks, merge rules, deduplication, and replay semantics.
When to Use
- implementing CDC ingestion
- designing watermark-based or timestamp-based incremental loads
- merging inserts, updates, and deletes
- handling late-arriving data
- fixing duplicate or missing incremental records
Do not use this when a bounded full refresh is simpler and operationally safer.
Workflow
-
Define the change contract. Include:
- business key
- change key or sequence
- delete behavior
- update semantics
- source ordering guarantees
-
Choose the incremental strategy. Typical options:
- append-only
- merge/upsert
- snapshot replacement
- log-based CDC
-
Make watermark logic explicit. Record:
- watermark column
- lag tolerance
- reprocessing window
- retry and replay rules
-
Guard against duplicates and missed changes. Deduplication and late-data handling should be first-class logic.
-
Prove reruns are safe. Incremental pipelines that cannot be replayed safely are incomplete.
Common Rationalizations
| Rationalization | Reality |
|---|---|
| "We can just use updated_at as the watermark." | Timestamps alone often miss late changes, clock skew, or delete semantics. |
| "Duplicates are rare enough." | Rare duplicates are still expensive when they affect published metrics. |
| "We can rebuild everything if something goes wrong." | Large backfills are often slow, expensive, or operationally risky. |
Red Flags
- no documented business key
- delete handling is absent
- watermark logic depends on assumptions no one wrote down
- replay requires manual surgery in production tables
Verification
- The incremental contract covers inserts, updates, and deletes where relevant
- Watermark, lag, deduplication, and replay rules are explicit
- Late data behavior is defined
- Reruns and backfills are operationally safe
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.
- 8d ago First seen · 73 lines · 48 tokens per session scan A 29f21c14a0de
cdc-and-incremental-loading is a skill published in the GitHub repository vaquarkhan/data-engineering-agent-skills (43 stars, last pushed 2mo ago), licensed MIT. It adds 48 tokens to every session and 507 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
cqrs-implementation
Implement Command Query Responsibility Segregation for scalable architectures. Use when separating read and write models, optimizing query performance, or building event-sourced systems.
projection-patterns
Build read models and projections from event streams. Use when implementing CQRS read sides, building materialized views, or optimizing query performance in event-sourced systems.
sl
Skill "sl" from Kaelio/ktx, covering semantic layer, part 1 - schema reference, overlay sources, standalone table sources and standalone sql sources.
historic_sql_table_digest
Convert one changed historic-SQL table usage bucket into typed table usage evidence for deterministic schema projection.
live_database_ingest
Capture semantic-layer and knowledge updates from a live database schema snapshot.
architect/data-api-design
A guide to designing data models and application programming interfaces (APIs), which are the rules software uses to exchange data.