Stack Palantir engineering context into Zyra agents and copilot

[sonoxo]

Adds a clean-room Palantir-inspired engineering stack for Zyra based on the user-supplied Palantir Learn Data Engineer and Application Developer guide links.

Includes:

• shared server/engineering-context.ts with stages, decision loop, and bounded specialist fleet
• tests for source provenance, fleet selection, approval gates, and context rendering
• live ZyraCopilot vision-model system context injection through the Hugging Face call path
• engineering fleet skill for agent workflows
• updates to adaptive-intelligence and architecture skills
• documentation for clean-room data/application engineering patterns

No Palantir guide text is copied verbatim and no Palantir affiliation, tenant access, certification, proprietary APIs, or implementation details are claimed.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: dbf1849325

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Add the required skill frontmatter

Add YAML frontmatter containing at least name and description. The skill loader uses those fields to discover and select skills, so without them zyra-engineering-fleet is omitted from the available skill registry and will not trigger for the engineering tasks this commit intends it to handle; references to it from the architecture and adaptive-intelligence skills do not restore independent discovery.

Useful? React with 👍 / 👎.