Cohere Data Handling and Privacy
SkillDev toolsGovern data sent to Cohere with classification, minimization, tenant controls, redaction, retention mapping, and contractual evidence. Use when processing sensitive or regulated data. Trigger with "Cohere privacy", "Cohere PII", or "Cohere data retention".
Use Cohere Data Handling and Privacy in Claude, ChatGPT or Ahel Desktop
Free. Sign in, add Cohere Data Handling and Privacy and connect your AI. About a minute.
Also: Claude Code · Cursor · Codex
Then ask your AI: use the Cohere Data Handling and Privacy skill
Details
Instructions available. Your AI can read the instructions. Execution depends on the setup they require.
Account requirements not reviewed. Check the skill instructions before use; Ahel provides instructions and does not run this skill.
No other account needed.
Add Ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
What this skill tells your AI
The instructions your AI receives, as published by jeremylongshore/tons-of-skills-marketplace in skills/.curated/cohere-data-handling/SKILL.md and read by Ahel’s review.
Overview
Decide what may cross the provider boundary from the signed agreement and application purpose, then enforce that decision before each request.
Prerequisites
- The target repository, runtime, environment, and accountable owner
- An approved Cohere team and key for any live verification
- Current quality, security, privacy, capacity, and change-control requirements
Tool Discipline
Use Read, Glob, and Grep to inspect code, configuration, and evidence. Use WebFetch only for current Cohere primary documentation. Use Write or Edit only when the user requested implementation and the exact target files are known; never write credentials or customer content.
Current Contract
- Cohere's public privacy policy distinguishes enterprise Customer Data from general personal information and points to enterprise commitments.
- Do not infer retention, residency, training use, deletion, or subprocessors from a generic policy when a contract governs the workload.
- Embeddings and logs can encode sensitive information even when raw text is absent.
- Data minimization and tenant authorization must happen before Chat, Embed, Rerank, or tool calls.
Authentication
Use an environment-specific key injected from an approved secret manager. Never print, persist, commit, or place CO_API_KEY in an example. Confirm access with the least costly bounded operation appropriate to the task, and treat key creation, rotation, revocation, role changes, and production-capacity requests as owner-approved actions.
Instructions
- Classify inputs, retrieved documents, outputs, embeddings, logs, and tool data by tenant and jurisdiction.
- Map each data class to the signed terms, approved region, retention, deletion, access, and incident requirements.
- Block disallowed classes and minimize, tokenize, or redact approved inputs before the provider call.
- Apply tenant filters before retrieval and keep document identifiers safe for citation auditing.
- Configure logs and caches with allowlisted fields, encryption, access control, and deletion tests.
- Produce a data-flow record and obtain privacy or legal approval for unresolved contractual claims.
Approval Boundaries
Do not expose or rotate keys, change Cohere Team roles, accept commercial terms, enable sensitive production data, increase spend or capacity, switch production models, send a support bundle, or execute model-proposed side effects without the accountable owner's approval. Keep diagnosis read-only unless implementation was requested.
Output
Return the resolved API and model contract, files or settings inspected, evidence collected, validation result, remaining risk, owner, and rollback or next action. Redact keys, authorization headers, prompts, retrieved documents, embeddings, customer identifiers, and unrestricted environment output.
Error Handling
| Condition | Response |
|---|---|
| Terms unavailable | Block sensitive production use until the governing agreement is identified. |
| PII detected | Apply the approved transformation or reject the request. |
| Cross-tenant data | Stop processing and invoke the data incident procedure. |
| Deletion unverified | Do not claim compliance; test each controlled storage layer. |
Examples
Use this compact handoff shape to keep the selected scope, validation evidence, and operational result reviewable.
Input:
data=customer-support; class=confidential; jurisdiction=EU; contract=enterprise
Expected handoff:
purpose=approved; fields=minimized; retention=mapped; review=recorded
Resources
Signals
- GitHub stars
- 3k
- Forks
- 415
- Last commit
- Oct 2026
Advanced
- Item type
- skill
- Key
cohere-data-handling- Source
- github.com/jeremylongshore/tons-of-skills-marketplace