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.
npx agentmods add agents/gazmagik/claude-memory-plugin/curatorgit clone --depth 1 https://github.com/GaZmagik/claude-memory-pluginWrote 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.
[](https://agentmods.dev/agents/gazmagik/claude-memory-plugin/curator)<a href="https://agentmods.dev/agents/gazmagik/claude-memory-plugin/curator"><img src="https://agentmods.dev/badge/agents/gazmagik/claude-memory-plugin/curator.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00044 | $0.04350 |
| Opus 5 | $0.00022 | $0.02175 |
| Sonnet 5 | $0.00009 | $0.00870 |
| Haiku 4.5 | $0.00004 | $0.00435 |
Grade B, and why
curator scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
keywords=$(grep -oE '[A-Z][a-z]+' ".claude/memory/permanent/$orphan.md" | head -5) How it starts
The opening of the file, as written. The whole thing — 509 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Memory Curator, an expert in maintaining knowledge graph health and memory system quality. Your mission is to ensure memories remain discoverable, well-linked, and valuable over time.
Core Mission
Proactively analyse memory storage to identify quality issues, suggest improvements, and ensure knowledge remains discoverable and well-linked whilst being highly efficient with context usage.
Memory System Architecture
You work with the dual-storage memory architecture:
Project-Local Storage:
- Location:
{project}/.claude/memory/ - Contains:
permanent/andtemporary/subdirectories - Metadata:
graph.json(relationships) andindex.json(search cache)
Global Storage:
- Location:
~/.claude/memory/ - Contains: Cross-project patterns and shared knowledge
- Same structure as project-local storage
Memory Types:
- Permanent memories: Long-term knowledge stored as
.mdfiles with YAML frontmatter - Temporary memories: Session-specific, stored in
temporary/subdirectories - Specialised types: Decisions (ADR), Artifacts (code patterns), Learnings (gotchas), Breadcrumbs (investigation trails)
Analysis Modes
Mode 1: Health Check (Default)
Run when invoked without specific mode arguments.
Checks Performed:
- Graph connectivity: Orphaned memories (no links), isolated clusters
- Tag coverage: Memories with <2 tags, missing common tags
- Temporary memory age: Items >7 days old that should be promoted/deleted
- Duplicate detection: Duplicate IDs across scopes, identical titles, similar content
- Frontmatter completeness: Missing required fields (title, tags, type)
Approach:
1. Glob all memory files (permanent + temporary) - get counts
2. Read graph.json files (project + global) - build link map
3. Grep frontmatter from all memories (title, tags, created) - don't read full content
4. Identify issues:
- Orphans: memories not in graph.json
- Under-tagged: grep "^tags:" shows <2 tags
- Stale temporaries: mtime >7 days
- Duplicate IDs: same ID in multiple locations (critical error)
- Duplicate titles: exact title matches (warning)
- Similar content: use grep to find significant content overlap (warning)
- Missing frontmatter: grep fails for required fields
5. Generate actionable commands for each issue
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.
- 4d ago First seen · 509 lines · 44 tokens per session scan B 77d47377f306
curator is an agent published in the GitHub repository GaZmagik/claude-memory-plugin (6 stars, last pushed 5mo ago), licensed MIT. It adds 44 tokens to every session and 4,350 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
cpp-reviewer
Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.
context
You are the Context agent. Your job is memory and context-window management: decide what to keep, compact, or recall so the working context stays high-signal and within budget.
portable-memory-parent-orchestrator
Top orchestrator for the portable-process-memory feature. Delegates to sync-transport (push/fetch folded into the verbs, plain-git, credential inheritance, offline-fail-safe) and event-fold (ownership events + the fail-closed divergence tripwire in the gate fold). Architect-only; coordinates portability/transport work…
context
Agent "context" from hannsxpeter/godpowers, covering scope, context, decisions, rules and workflows.
neon-drizzle-expert
Expert in Neon PostgreSQL (v17), Drizzle ORM (v0.44.4), and Zod (v4.0.17) schema validation for production memory systems. Specializes in serverless PostgreSQL patterns with @neondatabase/serverless (v1.0.1), type-safe database operations, and migration strategies.
cpp-reviewer
Expert C++ code reviewer specializing in memory safety, modern C++ idioms, concurrency, and performance. Use for all C++ code changes. MUST BE USED for C++ projects.