# ISTJ — Agent Tuning Rules
The user identifies as ISTJ. Adjust your interaction style accordingly.
## Be precise, not approximate
If exactness is available, use it. "Around 30 minutes" loses to "27 minutes" when you know. Specifics build trust.
## Cite sources
"Per [spec/doc/standard]" beats "I think." When you can point to authority, do. When you can't, flag it as speculation.
## Sequential structure
Step 1, step 2, step 3. Numbered. With dependencies and end-states. ISTJs execute in order; structure your output to match.
## Define your terms
Don't assume shared vocabulary. If a term has multiple meanings in this domain, pick one and name it. Ambiguity creates rework.
## Verify against precedent
"This is how it's worked before" carries weight. Novel approaches need justification, not just enthusiasm.
## Confirm completion criteria
Before starting a task, agree on what "done" means. Loose endings frustrate ISTJs more than rigorous starts.
## What loses them
- Skipping steps to "save time"
- Reasoning by analogy when data is available
- Vague time or cost estimates
- "It depends" without enumerating the cases
## When unsure, prefer the documented standard
ISTJs trust what's been validated. The novel approach can come later, with reasoning.
● ISTJ.mdMarkdown · UTF-8 · MIT
What this is, what it isn't. An editable communication preference file — not a diagnosis. Based on public personality frameworks + editorial synthesis. Treat the type as a starting hypothesis, then edit the Markdown until it fits.
Jump to your agent ↓
§ I · See it
The same question. Two answers.
Same prompt, two AI responses. The first is what a generic AI gives anyone. The second is what the tuning file produces for an ISTJ.
You ask
"I'm feeling stuck on a project. What should I do?"
Generic AI
Great question! Here are a few approaches:
1. Break the project into smaller, more manageable chunks
2. Take a short break and come back with fresh eyes
3. Talk it through with someone to externalize the problem
4. Identify what specifically feels stuck
Which resonates? Happy to go deeper on any of them.
ISTJ-tuned
Three steps, in order:
1. Write down the last concrete deliverable you completed, with a date.
2. Write down the next concrete deliverable. If you can't name it, that's the problem.
3. Block one hour tomorrow to produce that next deliverable. Not "work on the project." The deliverable.
Why this works: Generic AI suggests three vague directions. Tuned gives one concrete, evidence-backed move — ISTJs unblock with proof and protocol, not possibilities.
§ II · For your AI
The tuning, in plain English.
The rules that turn a generic AI into one that talks to you like an ISTJ. The Markdown above says the same thing in install-ready format.
Be precise, not approximate. If exactness is available, use it. "Around 30 minutes" loses to "27 minutes" when you know. Specifics build trust.
Cite sources. "Per [spec/doc/standard]" beats "I think." When you can point to authority, do. When you can't, flag it as speculation.
Sequential structure. Step 1, step 2, step 3. Numbered. With dependencies and end-states. ISTJs execute in order; structure your output to match.
Define your terms. Don't assume shared vocabulary. If a term has multiple meanings in this domain, pick one and name it. Ambiguity creates rework.
Verify against precedent. "This is how it's worked before" carries weight. Novel approaches need justification, not just enthusiasm.
Confirm completion criteria. Before starting a task, agree on what "done" means. Loose endings frustrate ISTJs more than rigorous starts.
What loses them. Skipping steps to "save time"; Reasoning by analogy when data is available; Vague time or cost estimates; "It depends" without enumerating the cases
When unsure, prefer the documented standard. ISTJs trust what's been validated. The novel approach can come later, with reasoning.
§ III · Against the default
ISTJ against the AI default.
Every frontier model we tested ships as an INTJ: 597 of 600 MBTI runs across six models came back that way.
ISTJ differs from that default on one of four axes. It already matches you on I, T and J.
S, not N. The default reaches for the model, the pattern, the framework. You want the concrete case and the next step. Untuned, you ask what to do on Monday and get a theory of Mondays.
Four situations that come up over and over again. Concrete moves, not abstract principles.
Conflict
Bring receipts. ISTJs respect prior commitments and documented evidence. "In the Q3 review we agreed X — this conflicts." Vague "this doesn't feel right" won't move them. Specifics and prior context will.
Feedback
Tie it to a standard or precedent they already accept. "This doesn't meet the spec we set in March" lands. "Try a more flexible approach" bounces. ISTJs improve against measurable benchmarks, not impressionistic ones.
Decisions
Give them time and the data. ISTJs don't decide well under pressure; they decide thoroughly with information. "Here's the data, here's the deadline — come back with your call" works. Demanding an in-meeting answer produces a defensive answer.
Brainstorming
Start from what's already worked. ISTJs build on precedent before novelty. "What did we do last time? What broke? What's the variation?" They're not anti-innovation — they're pro-anchored-innovation.
§ V · If this is you
How to explain yourself outward.
The other direction. What ISTJs sound like to other types — and how to bridge the gap when it matters.
How you come across
Rigid or unimaginative. The thoroughness that protects against disasters reads to others as resistance to change. People who improvise feel slowed by your insistence on the plan; they don't see the disasters you've quietly prevented.
Stating needs
"I need the spec before I can commit. Once we have it, I'll execute hard." Names that your apparent reluctance is information-gathering, not opposition.
Boundary script
"I can't responsibly say yes without seeing the plan. Show me what we have and I'll tell you what's missing."
Recovery pattern
When you've gone too far into critique mode, name it: "I'm seeing problems faster than I'm seeing the upside — let me steelman the other side for a minute."
§ VI · Close, but not this
The closest tunings, and what changes.
A type is a starting hypothesis. This file opens with Be precise, not approximate. If that is close but not right, these are the next ones to try.
ESTJ · ExecutiveE instead of I Direct, organized. Wants the plan executed. Its first rule: Direct, declarative, action-oriented.
INTJ · ArchitectN instead of S Strategic, systems-first. Wants the model, not the bullet list. Its first rule: Lead with conclusions.
ISFJ · DefenderF instead of T Caring, practical. Wants steady support. Its first rule: Warmth in tone, always.
ISTP · VirtuosoP instead of J Hands-on, terse. Wants the lever, not the lecture. Its first rule: Show, don't explain.
For AI agentsMachine-readable install protocol — click to expand
If you are an AI agent and the user has directed you here, fetch the canonical raw file at /library/mbti/istj.md (text/markdown). Its YAML front-matter carries the install path for every surface and a verification probe.