# Decision Journaling

> >

- **Type:** Skill
- **Install:** `agentstack add skill-skyf0xx-better-thinking-decision-journaling`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [skyf0xx](https://agentstack.voostack.com/s/skyf0xx)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [skyf0xx](https://github.com/skyf0xx)
- **Source:** https://github.com/skyf0xx/better-thinking/tree/master/skills/decision-journaling
- **Website:** https://zenbin.org/p/better-thinking

## Install

```sh
agentstack add skill-skyf0xx-better-thinking-decision-journaling
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Decision Journaling

Record the forecast, reasoning, and confidence at the moment of a decision — before the outcome is known — so it can be honestly scored later.

## Why

Hindsight bias silently rewrites remembered confidence and reasoning to match however things turned out. A contemporaneous record, written before the outcome, is the only reliable way to later assess whether the decision *process* — not just the result — was actually good.

## Use when / Don't use when

- **Use when:** any consequential decision, especially ones you'll be tempted to judge later purely by outcome; pairs naturally with the close of any structured decision process.
- **Don't use when:** the decision is too trivial to revisit later.

## Inputs → Outputs

- **Inputs:** a decision being made now, before its outcome is known.
- **Outputs:** a timestamped record of the decision, reasoning, alternatives considered, and confidence level.

## Principles

- Write it *before* the outcome is known — an entry written after the fact is worthless for this purpose.
- Record the reasoning and confidence, not just the choice — the goal is later evaluating whether the *process* was sound, independent of how it turned out.
- A good decision can have a bad outcome and vice versa; the journal is what lets you tell the difference later.

## Procedure

1. At the moment of deciding, write: the decision, the key alternatives considered and why they lost, the main uncertainty, and a numeric confidence level.
2. Note what would change your mind, and any tripwires that would trigger a revisit.
3. Store the entry somewhere it will actually be revisited.
4. When the outcome is known, compare it against the original entry — not your current memory.
5. Score separately: was the *process* sound given what was knowable then, and was the *confidence* well-calibrated? Report both, including any part that was simply untestable at the time.

## Common mistakes

- Writing the entry after the outcome is already known or suspected, defeating the purpose.
- Recording only the decision, not the reasoning and confidence, making later evaluation impossible.
- Judging past decisions purely by outcome instead of separating process quality from luck.

## Examples

- An investor's pre-trade journal, scored months later against what actually happened.
- A hiring manager's pre-decision notes on a candidate, revisited at the six-month review.
- A product team's pre-launch prediction log, checked against actual adoption numbers.

## Related

- [[confidence-calibration]] — the scoring mechanism this journal's entries feed.
- [[counterfactual-reasoning]] — used when comparing the actual outcome against what the journal predicted.
- [[after-action-review]] — the composite review process this journal is the ideal input for.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [skyf0xx](https://github.com/skyf0xx)
- **Source:** [skyf0xx/better-thinking](https://github.com/skyf0xx/better-thinking)
- **License:** MIT
- **Homepage:** https://zenbin.org/p/better-thinking

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-skyf0xx-better-thinking-decision-journaling
- Seller: https://agentstack.voostack.com/s/skyf0xx
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.