memory-prune

memory-prune is a command for Claude Code from smrafiz/claude-supercharger. It costs 0 tokens per session (448 once invoked), scanned A, original, MIT.

A command that archives resolved entries from Claude Code's project memory so they stop loading into every session.

In plain words
What is it for?
Use it first to preview entries that can be archived, then apply the archive to entries explicitly marked resolved or superseded and belonging to the project.
Why use it?
It reduces repeated context from finished, replaced, or disproved notes without deleting them.

Command for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable. Also seen: reads .claude/ paths.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the supercharger plugin — 37 commands, 10 agents, 28 hooks shipped together

Good fit Use it first to preview entries that can be archived, then apply the archive to entries explicitly marked resolved or superseded and belonging to the project.

Compare 6 commands from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add smrafiz/claude-supercharger
Claude Code
/plugin install supercharger

Made for: Claude Code.

Or install supercharger, the plugin that ships this one along with the rest of its 37 commands, 10 agents, 28 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for memory-prune

README.md
[![agentmods](https://agentmods.dev/badge/commands/smrafiz/claude-supercharger/memory-prune.svg)](https://agentmods.dev/commands/smrafiz/claude-supercharger/memory-prune)
Your own site
<a href="https://agentmods.dev/commands/smrafiz/claude-supercharger/memory-prune"><img src="https://agentmods.dev/badge/commands/smrafiz/claude-supercharger/memory-prune.svg" alt="Measured on agentmods" height="20"></a>
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 448 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00000 $0.00448
Opus 5 $0.00000 $0.00224
Sonnet 5 $0.00000 $0.00090
Haiku 4.5 $0.00000 $0.00045

Measured 8d ago against content hash 3d17ca6325aa, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

memory-prune 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 8d 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.

commands/memory-prune.md · 37 lines

What it actually says

Archive resolved memory entries so they stop loading into context every session. Arguments: $ARGUMENTS

Claude's file-memory (~/.claude/projects/<project>/memory/) loads its MEMORY.md index into context at the start of every session. Entries that are done — resolved, superseded, debunked — keep costing tokens forever. This prunes them safely: it only auto-archives entries the author explicitly marked status: resolved (or superseded) with type: project; everything else is only suggested. Nothing is deleted — entries move to memory/archive/ and can be restored.

Step 1 — show what would happen (default, no changes):

bash ${CLAUDE_PLUGIN_ROOT}/tools/memory-prune.sh

This prints three groups: Auto-archivable (will move on --apply), Suggestions (project entries with a terminal marker or >90d untouched — you decide), and entries marked resolved but not type: project (left in place by design).

Step 2 — decide. If $ARGUMENTS contains apply (or the user confirms), archive the auto-archivable set:

bash ${CLAUDE_PLUGIN_ROOT}/tools/memory-prune.sh --apply

For an entry in the Suggestions list that the user confirms is done, don't force it — edit that entry's frontmatter to add status: resolved under metadata:, then re-run --apply. That keeps the author-declared safety model intact.

Other actions:

  • bash ${CLAUDE_PLUGIN_ROOT}/tools/memory-prune.sh --list-archive — show archived entries.
  • bash ${CLAUDE_PLUGIN_ROOT}/tools/memory-prune.sh --restore <name> — bring one back (re-adds its index line).

Report the tool output verbatim, then state how many entries were archived (or would be) and the approximate per-session token saving (each index line ≈ 40–60 tokens).

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. 8d ago First seen · 37 lines · 0 tokens per session scan A 3d17ca6325aa

Subscribe to this mod's changes

memory-prune is a command published in the GitHub repository smrafiz/claude-supercharger (11 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 448 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.