Personal AI Agent Setup Guide

Personal AI agent setup worksheet

My first 2-4 workflows:
-
-
-
-

My durable source of truth should be:
- Local folder/repo:
- Existing tools it may read:
- Existing tools it may write to:
- Context that must stay private/local-only:

My first interface:
- Telegram / terminal / web / other:

My safety rules:
- Agent may read:
- Agent may write:
- Agent may never do:
- Requires explicit approval:

My week-one success criteria:
-
-
-

# Secret names to create later. Keep real values out of prompts and docs.
OPENAI_API_KEY="paste-real-value-in-secret-manager-only"
ANTHROPIC_API_KEY="paste-real-value-in-secret-manager-only"
TELEGRAM_BOT_TOKEN="paste-real-value-in-secret-manager-only"
TELEGRAM_CHAT_ID="paste-real-value-in-secret-manager-only"
GITHUB_TOKEN="paste-real-value-in-secret-manager-only"
DATABASE_URL="paste-real-value-in-secret-manager-only"

# Railway example: paste real values only into your local terminal or Railway UI.
railway variables --set "OPENAI_API_KEY=<real value>"
railway variables --set "TELEGRAM_BOT_TOKEN=<real value>"
railway variables --set "TELEGRAM_CHAT_ID=<real value>"

# Safe verification: show variable names without printing values.
railway variables | sed 's/=.*$/=<hidden>/'

# Human-only: create local secrets file.
cp .env.example .env
chmod 600 .env

# Then edit .env locally. Do not commit it.
# Safe check before commit:
git status --short
git check-ignore .env

You are helping me design a personal AI agent operating model.

Goal: create the first practical version, not a maximal architecture.

Inputs I will provide:
- My first 2-4 workflows
- My preferred durable source of truth
- My first interaction channel
- Privacy/security boundaries
- Week-one success criteria

Your task:
1. Turn the inputs into a concise operating model.
2. Propose a folder/file structure for durable context.
3. Separate human-only decisions from agent-executable tasks.
4. List the minimum integrations needed for version one.
5. Identify risks, secrets, and permissions that require explicit approval.
6. Produce a step-by-step setup plan with verification checks.

Rules:
- Do not invent credentials.
- Do not ask me to paste secrets into chat.
- Mark secret-handling steps as human-only.
- Prefer a reversible week-one setup over a big-bang architecture.

Create a local personal-agent source-of-truth structure using Markdown files.

Constraints:
- Keep it simple and readable.
- Do not store secrets.
- Use personal/outcome/system/task/decision/interaction/event/routine categories.
- Add README files that explain what belongs in each folder.
- Add a CONTEXT_POLICY.md that classifies what should and should not be saved.
- Add a RESOLVER.md that explains where new notes should be routed.

Suggested top-level folders:
- personal/
- outcomes/
- systems/
- tasks/
- decisions/
- interactions/
- events/
- routines/
- inbox/needs-triage/

After creating files:
1. Print a tree view.
2. Explain how a new note should be routed.
3. List files that should be reviewed before any agent writes durable context.
4. Confirm no secrets were created or requested.

Help me install a personal AI agent runtime safely.

Before changing files:
1. Inspect the repo/folder and identify the stack.
2. List required commands and dependencies.
3. Identify where environment variables should be stored.
4. Identify any commands that require human approval.

Implementation rules:
- Do not print or commit secrets.
- Do not push code unless I explicitly approve.
- Do not modify production systems.
- Prefer small checkpoints if this is a git repo.
- After each major step, run a verification command and report the result.

Deliverables:
- installed runtime or clear blocker list;
- .env.example or config template with placeholders only;
- runbook for starting/stopping the agent;
- verification checklist;
- list of secret names required, with no secret values.

Connect the first interface for my personal AI agent.

Interface selected: <Telegram | terminal | web | other>

Rules:
- Human creates credentials and stores secrets.
- Agent may create config templates with placeholder names only.
- Agent may update docs/runbooks.
- Agent must not send external messages except explicit test messages I approve.

Tasks:
1. Inspect existing config and docs.
2. Add or update interface configuration using placeholders.
3. Add a runbook for local testing and hosted deployment if applicable.
4. Add a verification checklist with exact test messages or commands.
5. Report any missing credentials or manual setup steps.
6. Confirm no secret values were printed or committed.

Add first-week routines for my personal AI agent.

Candidate routines:
- daily planning review;
- weekly review;
- inbox triage;
- content idea capture;
- project status digest;
- follow-up reminders.

Rules:
- Start advisory-only unless I explicitly approve automation.
- No external sends without approval.
- No spending money.
- No deleting files.
- No production changes.
- Durable context must be routed according to CONTEXT_POLICY.md and RESOLVER.md.

Deliverables:
1. Routine definitions with cadence, inputs, outputs, and owner.
2. Safety gates for each routine.
3. Manual test procedure.
4. Rollback/disable procedure.
5. First-week review checklist.

Version-one personal agent acceptance checklist

[ ] Operating model written and reviewed
[ ] Durable source-of-truth folder/repo created
[ ] Context policy and resolver exist
[ ] .env.example or secret-name checklist exists
[ ] Real secrets stored outside prompts/docs
[ ] Runtime starts locally or in staging
[ ] First interface sends and receives a test message
[ ] At least one routine can run manually
[ ] Safety gates are documented
[ ] Human can disable the agent or routine quickly
[ ] Week-one review is scheduled