Add multi-LLM prompt techniques catalog and governance links

[2-Coatl]

Summary

• add PROMPT_TECHNIQUES_CATALOG.md with the full prompt engineering compendium and record the associated ExecPlan
• link the new catalog from the prompting index, README, provider/domain agent fichas, and SDLC guide so all LLM workflows reference the same source
• extend the documentation alignment tests to guard the catalog and update the agent governance directory accordingly

Testing

• pytest docs/testing/test_documentation_alignment.py

---

Codex Task

[chatgpt-codex-connector]

💡 Codex Review

https://github.com/2-Coatl/IACT---project/blob/dbca38b0c50fa2c58842148137986a8de0769bf1/scripts/coding/ai/shared/env_loader.py#L22-L24
Fix repo root detection when loading .env

The new .env loader never actually reaches the repository root, so it silently ignores the file it is supposed to read. In scripts/coding/ai/shared/env_loader.py you stop ascending at Path(__file__).parent.parent.parent.parent, which for this module resolves to /scripts, not /. As a result env_path points to scripts/.env, load_env_file() always returns {}, and get_llm_config_from_env()/verify_api_keys() will never see the API keys that live in the real project root. This regresses the very feature the module introduces (automatic provider detection). Please climb one more level before appending .env so the loader inspects the repository root instead of scripts/.

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".