brief-feedback

name	brief-feedback
description	Process feedback on morning briefs and update quality rules. Use when the brief was wrong or needs improvement.

Brief Feedback Skill

Process feedback on morning briefs and update quality rules.

Trigger

/brief-feedback followed by your feedback

What It Does

1. Takes your natural language feedback about the brief
2. Extracts actionable rules from the feedback
3. Appends new rules to Vault/Core Memory/Brief Feedback.md
4. Next brief will incorporate these rules

Example Usage

/brief-feedback The Alexei suggestion was bad - you shouldn't propose partnerships without checking their GitHub first. Also stop suggesting random apartment stuff as quick wins.

Atlas will:

1. Extract rules: "Check GitHub before suggesting partnerships", "Quick wins must relate to 2026 goals"
2. Add to Brief Feedback.md under appropriate sections
3. Confirm what was added

Execution Steps

When user invokes /brief-feedback [feedback]:

1. Read current Vault/Core Memory/Brief Feedback.md
2. Analyze the user's feedback to extract:
   • New "Don't Do" rules
   • New "Do More" rules
   • New "Bad → Good" examples
   • New context that matters
3. Append extracted rules to appropriate sections
4. Update the "Last updated" date
5. Confirm changes to user

Rule Extraction Guidelines

From feedback, extract:

Feedback Pattern	Rule Type
"Don't suggest X" / "Stop doing Y"	Don't Do
"You should X first" / "Always Y before Z"	Do More
"X was bad, should have been Y"	Bad → Good Example
"I care about X" / "X is important"	Context That Matters
"X is higher priority than Y"	High-Priority Situations

Output Format

After processing feedback:

✅ Added to Brief Feedback:

**Don't Do:**
- [new rule]

**Bad → Good:**
| [bad pattern] | [good pattern] |

Updated: Vault/Core Memory/Brief Feedback.md