Skip to content

[Codex] Add candidate Relationship Effectiveness templates #132

Description

@sergstack

Goal

Add candidate Relationship Effectiveness templates without creating a CRM project, outreach pipeline, or automation.

Context

This is a lightweight personal effectiveness layer connected to Weekly AI-OS Review, Communication Pack, and Productivity Operating Layer. It must remain candidate and human-reviewed.

Pre-check performed in chat: searched open issues for relationship networking outreach meeting recap personal CRM contacts follow-up; no direct duplicate was found. Codex should still inspect repo files and issue context before editing.

First step

Search existing open issues and repo files for:

  • relationship
  • networking
  • outreach
  • meeting recap
  • personal CRM
  • contacts
  • follow-up

If overlap exists, report it and do not duplicate.

Scope

Add minimal candidate templates/checklists only, if repo conventions support the destination.

Candidate files, subject to repo convention check:

  • RELATIONSHIP_CRM_LITE_TEMPLATE.md
  • WEEKLY_RELATIONSHIP_REVIEW_BLOCK.md
  • VALUE_FIRST_OUTREACH_TEMPLATE.md
  • MEETING_RECAP_TEMPLATE.md
  • ASK_FOR_ADVICE_TEMPLATE.md
  • NO_SPAM_HUMAN_REVIEW_RULE.md

Candidate destination, subject to inspection:

  • ChatGPT/[LLM]/Templates/
  • ChatGPT/[AI OS]/Knowledge/
  • ChatGPT/[Thinking]/Templates/
  • or existing equivalent repo convention.

Requirements

  • Templates remain candidate / ready for human review.
  • No production promotion without pilot evidence and owner acceptance.
  • All outbound messages require human review.
  • No auto-send.
  • No mass outreach.
  • No autonomous follow-up.
  • Real contacts and private notes must not be stored in public repo templates.
  • Repo examples must be empty or synthetic.
  • Prompt QA required before promoting outreach prompts or Stream Deck commands.
  • Stream Deck commands remain candidate until tested and accepted.

Template coverage

Relationship CRM Lite template should include empty fields only, such as:

  • name
  • role
  • organization
  • relationship type
  • topic of mutual interest
  • last contact date
  • last context
  • promised follow-up
  • value I can offer
  • ask / opportunity
  • next touch
  • status
  • notes

Weekly Relationship Review block should include:

  • who can benefit from current AI-OS / audit / finance artifact
  • who I owe a follow-up to
  • what value I can send this week
  • one relationship next action

Value-first Outreach template should require:

  • context
  • useful artifact / idea
  • why it may help
  • low-pressure next step
  • human review before sending

Meeting Recap template should include:

  • meeting
  • date
  • context
  • key points
  • commitments from me
  • commitments from them
  • useful follow-up
  • next touch
  • risks / sensitivities

No-spam / human-review rule should state:

  • no mass messaging
  • no auto-send
  • no autonomous follow-up
  • no fake praise
  • no private data in repo
  • human review before any outbound message

Forbidden

  • No CRM project.
  • No outreach pipeline.
  • No mass messaging.
  • No auto-send.
  • No autonomous follow-up.
  • No production automation.
  • No autonomous agents.
  • No autonomous retrieval.
  • No vector DB.
  • No embeddings.
  • No semantic search.
  • No web UI.
  • No real contact data.
  • No emails, phone numbers, addresses.
  • No private relationship notes.
  • No client/vendor/employee/bank/personal data.
  • No secrets or credentials.
  • No raw dumps.
  • No runtime artifacts.
  • No PROJECT_INSTRUCTIONS.md edits unless required and justified.
  • No auto-merge.

Checks to run

python3 scripts/check_project_instructions_length.py
python3 scripts/check_repo_public_safety.py
python3 scripts/check_manifest_paths.py
python3 scripts/check_knowledge_bundles.py
python3 scripts/check_codex_goal_mode_defaults.py
git diff --check

Acceptance criteria

  • Existing issue/repo overlap checked.
  • Minimal reversible diff.
  • Candidate templates added only if no duplicate scope exists.
  • Privacy/public repo boundary is explicit.
  • No auto-send / human-review rule is explicit.
  • No production promotion implied.
  • No blocked promotion items introduced.
  • No public repo safety risk introduced.
  • Upload/manifest references updated only if repo conventions require it.
  • If Knowledge bundles/upload lists change, note that live ChatGPT Project sync and smoke QA must be checked separately.
  • Checks pass or blockers are documented.
  • Rollback note included.
  • No auto-merge.

Stop conditions

Stop and report blockers if:

  • existing open issue or repo artifact already covers the same scope;
  • no clear destination folder convention exists;
  • work requires CRM project, outreach pipeline, mass messaging, auto-send, or autonomous follow-up;
  • work requires production automation or blocked promotion items;
  • work requires real contacts, private notes, emails, phone numbers, personal/client/vendor/employee data, secrets, raw dumps, or runtime artifacts;
  • work requires Stream Deck command promotion without Prompt QA;
  • work requires promotion to active without pilot evidence and owner acceptance.

Rollback

If PR is rejected or checks fail without a safe fix, revert only added Relationship Effectiveness templates and any related manifest/upload references.

Expected PR summary

  • Added candidate Relationship Effectiveness templates.
  • Existing overlap checked.
  • No CRM project.
  • No outreach pipeline.
  • No auto-send or mass messaging.
  • No real contact data.
  • No production automation.
  • No candidate promotion.
  • Checks listed.
  • Residual risks and rollback documented.
  • No auto-merge.

Metadata

Metadata

Assignees

No one assigned

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions