Delete, redact, or archive project and global memories when the user says forget this, delete memory, remove that preference, do not remember, redact sensitive information, or retire stale memory.
Scanned 9/11/2026
Install to Claude Code
npx -y skills add dvy1987/agent-loom --skill memory-forget --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Memory Forget?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/dvy1987-memory-forget)More formats (shields.io, HTML) on the badges page.
---
name: memory-forget
description: >
Delete, redact, or archive project and global memories when the user says
forget this, delete memory, remove that preference, do not remember, redact
sensitive information, or retire stale memory.
license: MIT
metadata:
author: dvy1987
version: "1.1"
category: project-specific
resources:
references:
- examples.md
---
# Memory Forget
You remove memory deliberately and traceably. Forgetting is a first-class memory capability, especially for global memory.
## Workflow
1. Identify exact target memory and scope: project, global, or both.
2. Confirm ambiguity if multiple entries match.
3. Classify action: delete, redact, archive, mark superseded, or add do-not-store rule.
4. For secrets or sensitive data, redact immediately and report the affected files.
5. Update index and routing files so forgotten memory is not recalled.
6. If archiving, move entry to `archived/` with reason and date.
7. If global memory shrinks, update `global-index.md`.
8. Log project file changes in `docs/skill-outputs/SKILL-OUTPUTS.md`.
## Forget Actions
| Request | Action |
|---|---|
| "forget this preference" | delete or mark retired in global memory |
| "do not remember X" | add a do-not-store rule if useful |
| "remove old decision" | mark superseded or archive, do not erase rationale unless requested |
| "this contains a secret" | redact and invoke security review |
## Hard Rules
- User deletion requests override convenience.
- Do not keep active index references to forgotten entries.
- Do not silently delete decision rationale when archival is safer.
- Do not archive secrets; redact them.
## Output Format
```markdown
Forget complete
Target: <memory>
Scope: <project/global/both>
Action: delete | redact | archive | retire
Files changed: <paths>
Residual references removed: yes/no
```
## Example
User: "Forget that global preference; it is no longer true."
Output: remove or retire the matching entry in `~/.agent-loom/memories/user-preferences.md` and update `global-index.md`.
## Common Rationalizations
| Excuse | Reality |
|--------|---------|
| Skip memory — just code | Next agent loses decisions, blockers, and approved scope. |
| Load every memory file | Read indexes and handoff tail only — bounded context. |
| Global memory for everything | Project memory default; global only when stable and cross-project. |
| External paste → memory | Run secure-* first; transform to agent-authored notes. |
## Verification
- [ ] Correct sub-skill routed with reason
- [ ] No secrets or raw transcripts persisted
- [ ] Files changed listed in Impact Report
- [ ] Security gate noted when external content involved
## Red Flags
- User deletion request overridden for convenience
- Index still references forgotten entry paths
- Decision rationale deleted when archival was safer
- Forget run without checking downstream handoff refs
## Prune Log
Last pruned: 2026-07-04
- No changes — citation audit passed; content current (improve-skills full pass 2026-07-04)
## Impact Report
After completing, report:
```markdown
Memory forgotten: <target>
Action: <delete/redact/archive/retire>
Scope: <scope>
Indexes updated: yes/no
Security issue: yes/no
```
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!