memory-forget

A command for marking a topic as forgotten in managed memory. It supports soft forgetting and, when explicitly requested, stronger removal modes.

In plain words
What is it for?
Use it to preview and apply forgetting for a topic, choose between soft, hard, and purge modes, and report exactly what each mode affects.
Why use it?
It clarifies what will be removed before changing memory and makes the limits of forgetting explicit; it does not erase copies in Git history, backups, or forks.

Command

Part of the memory-custodian plugin — 1 skill, 3 commands, 1 hook shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/waittim/memorycustodian/memory-forget
Clone the repo
git clone --depth 1 https://github.com/waittim/MemoryCustodian

Or install memory-custodian, the plugin that ships this one along with the rest of its 1 skill, 3 commands, 1 hook.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 173 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00173
Opus 5 $0.00000 $0.00086
Sonnet 5 $0.00000 $0.00035
Haiku 4.5 $0.00000 $0.00017

Measured 3d ago against content hash 2cc386f4a706, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

memory-forget scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

adapters/claude-code/commands/memory-forget.md · 16 lines

What it actually says

memory-forget

Run:

memory-custodian forget "<topic>" --mode soft
# Review the preview, then apply its Plan ID:
memory-custodian forget "<topic>" --mode soft --apply --confirm-plan <PLAN_ID>

Use --mode hard or --mode purge only when explicitly requested. Hard upgrades prior topic-bearing soft tombstones to a generic guard; purge removes them. If preview reports Manual rewrite required, rewrite the body or preamble semantically before applying.

Report the boundary exactly as shown by the preview: hard removes matching active managed memory, while purge also targets managed archive/. Neither mode rewrites Git history, commits the working tree, or revokes clones, forks, backups, caches, and other distributed copies. Never describe either result as permanent deletion everywhere.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 16 lines · 0 tokens per session scan A 2cc386f4a706

Subscribe to this mod's changes

memory-forget is a command published in the GitHub repository waittim/MemoryCustodian (20 stars, last pushed 4d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 173 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.