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 agentmods add skills/collibra/chip/data-product-createnpx skills add collibra/chip --skill data-product-creategit clone --depth 1 https://github.com/collibra/chipWhat 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 | $0.00058 | $0.04261 |
| Opus 5 | $0.00029 | $0.02131 |
| Sonnet 5 | $0.00012 | $0.00852 |
| Haiku 4.5 | $0.00006 | $0.00426 |
Grade A, and why
data-product-create 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 3d 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 — 279 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Creating a Data Product from a table
Packages an existing physical table — given as a name, DGC UUID, or Collibra URL — into a
Collibra Data Product with one Port and a Data Contract. Phases 1–5 are read-only discovery and a
proposal; Phase 6 is the single confirmation; Phase 7 writes. Generic tool mechanics — how to set
attributes, write Markdown, handle each tool's success/error responses, resolve relation names,
and recover from partial failures — live in collibra/asset-create and collibra/asset-edit;
this skill is the DP-specific layer on top.
Hard rules
- Confirm once. Do discovery silently, gather every required choice (included tables, domain, owner, SLAs) in as few turns as possible and before the overview, present one complete overview (Phase 5), take a single explicit approval (Phase 6), then execute all writes (Phase 7) with no further prompts. Never ask field-by-field; never re-confirm individual writes.
- Nothing is auto-included or auto-created without that approval. Adding a table is an access decision: suggested tables default to not included and the user opts them in. Creating assets is significant and not auto-rolled-back — wrong assets must be deleted manually.
- One Port by default; more only if the user explicitly asks.
- Fill every applicable attribute — the reference tables are the source of truth for which
attributes each asset has. Draft a proposed value for every attribute listed there; leave
blank only the SLAs the user declined and
Pricingwithout supplied cost. SLA values are discrete Data Contract attributes — never written intoDescriptionor the manifest. - Existing coverage is a stop condition: if a Data Product/contract already covers substantially the same tables, stop and offer to extend it instead.
- Say where inferred values come from. When a value is drawn from another asset or otherwise inferred, tell the user in chat where it came from so they can judge it — the written attribute holds only the clean value, not provenance markers.
- Attribute writes (per
collibra/asset-edit): set a not-yet-valued attribute withadd_attribute, notupdate_attribute(which only changes an existing value). If any name doesn't resolve on this instance, follow the error's suggestion.
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.
- 3d ago First seen · 279 lines · 58 tokens per session scan A 325041f72ead
data-product-create is a skill published in the GitHub repository collibra/chip (36 stars, last pushed 5d ago), licensed Apache-2.0. It adds 58 tokens to every session and 4,261 once invoked, about $0.0003 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
cognee-community
Use when the user needs something that ships outside cognee core — community database adapters (Qdrant, Milvus, Weaviate, Redis, Pinecone, FalkorDB, Memgraph, DuckDB, NetworkX, …), data-source connectors (Slack, Gmail, Notion, Confluence, Google Drive), custom tasks/pipelines/retrievers (Exa, ScrapeGraph, codify)…
bigquery-ai-ml
Skill for BigQuery AI and Machine Learning queries using standard SQL and AI. functions (preferred over dedicated tools).
cocoindex
This skill should be used when building data processing pipelines with CocoIndex, a Python library for incremental data transformation. Use when the task involves processing files/data into databases, creating vector embeddings, building knowledge graphs, ETL workflows, or any data pipeline requiring automatic change…
data-quality-frameworks
Implement data quality validation with Great Expectations, dbt tests, and data contracts. Use when building data quality pipelines, implementing validation rules, or establishing data contracts.
fabriciq-ontology-cli
Manages Fabric IQ Ontology items, including entity and relationship types, data bindings, and definition updates, plus schema, lineage, grounding, and graph-walk exploration. Use for ontology modelling and traversal. For natural-language questions over a Power BI report use fabriciq.
lineage-diff
Compare column-level lineage between two versions of a SQL query to show added, removed, and changed data flow edges.