home/lifestyle/instruction-health-report

Instruction Health Report

GPTClaudeGemini··102 copies·updated 2026-07-14
instruction-health-report.prompt
# Instruction Health Report

Date:
Scope:
Trigger:
Reviewer:

## Score

Instruction Health Score: __/100

| Dimension | Score | Notes |
|---|---:|---|
| Bloat risk |  |  |
| Duplicate guidance |  |  |
| Conflict risk |  |  |
| Stale rule load |  |  |
| Promotion discipline |  |  |
| Regeneration opportunity |  |  |

## Evidence Reviewed

- 

## Promotion Candidates

| Candidate | Destination | Evidence | Decision |
|---|---|---|---|
|  |  |  |  |

## Demotion Or Deletion Candidates

| Rule | Current surface | Reason | Proposed action |
|---|---|---|---|
|  |  |  |  |

## Autoimmune Findings

| Finding | Symptom | Harm | Recommendation |
|---|---|---|---|
|  |  |  |  |

## Regeneration Candidates

| Cluster | Replacement principle | Preserves | Removes |
|---|---|---|---|
|  |  |  |  |

## Memory Routing

| Finding | Route | Rationale |
|---|---|---|
|  |  |  |

## Human Approval Checklist

- [ ] No one-off issue was promoted directly.
- [ ] No duplicate/conflicting guidance was added.
- [ ] Every active-rule proposal has recurrence/severity evidence.
- [ ] Regeneration was considered before adding new active memory.
- [ ] Sensitive data and credentials were not copied into the report.

when to use it

Community prompt sourced from the open-source GitHub repo inbusiness23/agent-immune-system (MIT). A "Instruction Health Report" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.

tags

lifestylecommunitygeneral

source

inbusiness23/agent-immune-system · MIT