Getting it into your agent
There is no command for this one: it runs only inside a plugin, and the catalogue could not identify which plugin ships it. The source is linked below.
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.
[](https://agentmods.dev/instructions/tokligence/redmem/claude-md)<a href="https://agentmods.dev/instructions/tokligence/redmem/claude-md"><img src="https://agentmods.dev/badge/instructions/tokligence/redmem/claude-md.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 5.1 | $0.04890 | $0.04890 |
| Opus 5 | $0.02445 | $0.02445 |
| Sonnet 5 | $0.00978 | $0.00978 |
| Haiku 4.5 | $0.00489 | $0.00489 |
Grade A, and why
redmem CLAUDE.md 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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 385 lines — stays where its author put it; the contents beside it link to each section on GitHub.
redmem — development guide
Claude Code hooks: a secret shield, session memory, an image compressor, an autopilot loop, and a set of guards. Everything runs as a short-lived subprocess spawned per tool call, which shapes almost every constraint below.
Read docs/design/architecture.md for the full specification. This file is
what you need before touching the code.
The one that will bite you first: the interpreter
python3 on PATH is not necessarily the interpreter with your packages.
A hook process does not source your shell rc. So the python3 that resolves
correctly when you type it can resolve to something else entirely when Claude
Code spawns the hook. An alias python3=... is worse than useless here: an
alias exists only in interactive shells, so it hides the split rather than
fixing it.
On 2026-07-27 this destroyed a 235-entry encrypted vault twice in one session.
The machine had two brew pythons; the one holding the linked python3 had no
cryptography, so the shield could not decrypt its own mapping, treated it as
empty, and truncated it. Every check run by hand said the right interpreter.
Consequences for you:
- Run tests with an interpreter that has
cryptography, not with whateverpython3resolves to. Verify before you trust it:<your python3> -c 'import sys, cryptography; print(sys.executable)' install.shresolves an interpreter, requiresimport cryptography, and writes its absolute path intosettings.json. It refuses to install if none qualifies. Do not "simplify" that back topython3.- Claude Code caches hook commands at session start. Editing
settings.jsondoes not affect the running session. Afterinstall.sh, restart Claude Code or you will be testing the old configuration and drawing confident, wrong conclusions.
Running the tests
<python-with-cryptography> -m pytest -q # full suite, ~90s
<python-with-cryptography> -m pytest test_vault_integrity.py -q
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.
- 6d ago First seen · 385 lines · 4,890 tokens per session scan A fc8465bd2c63
redmem CLAUDE.md is an instructions file published in the GitHub repository tokligence/redmem (4 stars, last pushed 10d ago), licensed Apache-2.0. It adds 4,890 tokens to every session, about $0.0244 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 2026-08-31.
Other instructions, from other repositories
ggmcp AGENTS.md
AGENTS.md instructions for GitGuardian/ggmcp, covering technologies used, dependencies management, server implementation guidelines, code organization and imports and lower level python rules.
klaudiush CLAUDE.md
Claude Code instructions for smykla-skalski/klaudiush, covering claude.md, project overview, commands, completion (shell completion scripts) and doctor (diagnose setup and configuration).
vibeguard AGENTS.md
AGENTS.md instructions for majiayu000/vibeguard, covering agent instructions, scope, start here, core rules and delivery policy.
safedeps AGENTS.md
Instructions for aldegad/safedeps, covering agents.md — safedeps, engine support, architecture invariants (do not break), version ssot and docs.
phantom-secrets copilot-instructions.md
Copilot instructions for ashlrai/phantom-secrets, covering copilot instructions, key rules, commands, how it works and when you see api keys.
chronos AGENTS.md
AGENTS.md instructions for OthmanAdi/chronos, covering chronos: time awareness for agents, getting the facts, the seven rules and do not.