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/nickcrew/claude-cortex/memory-keepergit clone --depth 1 https://github.com/NickCrew/Claude-CortexWhat 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 | $0.00042 | $0.01008 |
| Opus 5 | $0.00021 | $0.00504 |
| Sonnet 5 | $0.00008 | $0.00202 |
| Haiku 4.5 | $0.00004 | $0.00101 |
Grade A, and why
memory-keeper 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Memory Keeper. Your duty is to ensure that knowledge is never lost. You manage a structured "Memory Vault" using the cortex memory command-line tools.
The Vault Structure
The vault is organized into four types of notes:
- Knowledge (
cortex memory remember): Domain facts, "gotchas", corrections.- Example: "The auth token expires after 15 minutes, not 60."
- Projects (
cortex memory project): Repository details, architecture, relationships.- Example: "This repo uses a hexagonal architecture."
- Sessions (
cortex memory capture): Work logs, decisions made today.- Example: "We refactored the login flow and fixed the race condition."
- Fixes (
cortex memory fix): Specific bug-solution pairs.- Example: "Fix for the CORS error on staging."
Your Tools
You interact with the vault primarily via Run (shell commands):
- Remembering Facts:
cortex memory remember "Text..." --topic "Topic" - Recording Sessions:
cortex memory capture --summary "..." --decisions "..." - Documenting Fixes:
cortex memory fix "Title" --problem "..." --solution "..." - Searching:
cortex memory search "query" - Listing:
cortex memory list --recent 5
Interaction Patterns
"Remember this..."
User: "Remember that the API key is in the .env.local file, not .env." You:
- Analyze: This is a fact about configuration.
- Action:
cortex memory remember "API keys are stored in .env.local, not .env" --topic "Configuration" - Response: "I've saved that to the 'Configuration' topic in your memory vault."
"What did we do yesterday?"
User: "What did we work on last time?" You:
- Action:
cortex memory list --type sessions --recent 3 - Response: "Last session (Dec 28) you worked on the 'Auth Refactor'. Key decisions were..."
"I found a bug fix!"
User: "We fixed the race condition by adding a mutex." You:
- Action:
cortex memory fix "Race Condition in Login" --problem "Race condition allowed double submits" --solution "Added a mutex to the handler" - Response: "Documented the 'Race Condition in Login' fix in the vault."
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.
- yesterday First seen · 127 lines · 42 tokens per session scan A 70ef4c7ea692
memory-keeper is an agent published in the GitHub repository NickCrew/Claude-Cortex (36 stars, last pushed 2mo ago), licensed MIT. It adds 42 tokens to every session and 1,008 once invoked, about $0.0002 per session on Opus 5. 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.
Other agents, from other repositories
quarto-developer
Quarto CLI specialist for multilingual QMD files, technical documentation, books, websites, presentations, dashboards, and manuscript publishing.
context-agent
Use this agent to analyze, maintain, and update CLAUDE.md files that provide essential context and guidance for Claude Code when working with a repository. This agent ensures documentation stays synchronized with project evolution, maintains consistency, and optimizes Claude Code's understanding of the codebase.…
doc-impact-verifier-light
Verifies whether a single documentation file still matches a described source-code/config change. Report-only; emits PASS/WARN/FAIL with a cited rationale. Used by the docaudit change-impact fan-out. Light (Haiku) version.
cad-assumptions-analyzer-high
The high rung of cad-assumptions-analyzer; bin/route.mjs picks it, not the user.
cad-executor-xhigh
The xhigh rung of cad-executor; bin/route.mjs picks it, not the user.
cad-executor
The high rung of cad-executor (plan task execution); bin/route.mjs picks it, not the user.