What is skill-memory-bank CLAUDE.md?

A project-memory system that stores rules, plans, research, status, and specifications in a `.memory-bank` folder. TDD, or test-driven development, means writing a failing test before implementing the code that makes it pass.

Use it to manage planned work, read project rules, preserve session knowledge, coordinate multiple working sessions, and carry out tasks through the memory-bank workflow.

It helps an agent recover project context between sessions and requires new logic to be checked with tests before and after implementation.

skill-memory-bank CLAUDE.md — Instructions file for coding agents by fockus: A project-memory system that stores rules, plans, research, sta

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 instructions/fockus/skill-memory-bank/claude-md

Clone the repo

git clone --depth 1 https://github.com/fockus/skill-memory-bank

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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/fockus/skill-memory-bank/claude-md.svg)](https://agentmods.dev/instructions/fockus/skill-memory-bank/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/fockus/skill-memory-bank/claude-md"><img src="https://agentmods.dev/badge/instructions/fockus/skill-memory-bank/claude-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 4,774 This file is loaded in full into every session.

When invoked 4,774 The same file — it is already loaded in full.

Security scan B 1 finding. 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.

Model	Per session	Once invoked
Fable 5	$0.04774	$0.04774
Opus 5	$0.02387	$0.02387
Sonnet 5	$0.00955	$0.00955
Haiku 4.5	$0.00477	$0.00477

Measured 5d ago against content hash 65ecd95bb9cd, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

skill-memory-bank CLAUDE.md 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 5d 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.

- AGR-025: memsearch-плагин выключен глобально (enabledPlugins=false): в MB-проектах дублирует /mb recall (session summaries + agreements/progress/notes), его Stop-хук гонял ONNX bge-m3 индексацию после каждого хода (сот

CLAUDE.md · 101 lines

How it starts

The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Memory Bank Skill

Long-term project memory through .memory-bank/, engineering rules, SDD specs, executable /mb work tasks, verification, review, and session persistence.

Hard Rules

Resolve the active Memory Bank before project work.
- Existing bank → print [MEMORY BANK: ACTIVE].
- No bank → print [MEMORY BANK: ABSENT]; do not initialize unless explicitly requested.
Read the project rules and Memory Bank context before implementation:
- global rules: rules/RULES.md from this skill bundle;
- project overrides: <repo>/AGENTS.md, <repo>/RULES.md or <bank>/RULES.md when present;
- core context: <bank>/status.md, checklist.md, roadmap.md, research.md when present (the resolver also detects legacy-cased layouts).
New logic requires TDD: failing test first, then implementation, then verification.
Do not bypass an existing plan/spec. If work comes from Memory Bank, execute through /mb work or the equivalent scripts.
If .memory-bank/COORDINATION.md exists, parallel sessions share the working tree: read the board before stages, commits, and shared-file edits; scoped git add only (never -A); obey FREEZE entries. Protocol: references/coordination.md.

Mandatory `/mb work` Gate

When a project has an active Memory Bank and the user says: implement, fix, continue, resume, next step, go by the plan, execute the spec, or similar:

Resolve workflow from <bank>/pipeline.yaml with scripts/mb-workflow.sh.
Resolve target/range with scripts/mb-work-resolve.sh and scripts/mb-work-plan.sh.
Treat specs/<topic>/tasks.md blocks marked  as executable source of truth.
If using a wrapper plan, it must have linked_spec or  markers. If not, stop and fix the wrapper before coding.
Follow resolved steps exactly. For governed workflows this means: implement → verify → review → judge → fix/backlog → done.
Pass exact model and thinking from pipeline.yaml/JSON lines to subagents. Do not use fuzzy model names.
Do not claim completion until configured verification/review/judge gates are satisfied, or the user explicitly chooses a simpler workflow.

Read the full file on GitHub · 101 lines

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.

5d ago First seen · 101 lines · 4,774 tokens per session scan B 65ecd95bb9cd

Subscribe to this mod's changes

skill-memory-bank CLAUDE.md is an instructions file published in the GitHub repository fockus/skill-memory-bank (25 stars, last pushed 1mo ago), licensed MIT. It adds 4,774 tokens to every session, about $0.0239 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 2026-08-30.

Memory and context Planning agent-skill ai-agents ai-coding-assistant anthropic claude claude-code

vitaecontext GEMINI.md

Gemini CLI instructions for vitaecontext/vitaecontext, covering vitaecontext, module map and runtime loading.

vitaecontext/vitaecontext · 355 tokens

LLM-Wiki AGENTS.md

Instructions for TrueHOOHA/LLM-Wiki, covering llm wiki schema, mission, okf conformance, non-negotiables and directory layout.

TrueHOOHA/LLM-Wiki · 4,245 tokens

Imprint GEMINI.md

Instructions for ilang-ai/Imprint: You have the Imprint extension active. Imprint learns how the user works and applies their preferences automatically.

ilang-ai/Imprint · 316 tokens

growmos AGENTS.md

Instructions for codician-team/growmos: This repository keeps a knowledge graph in .growmos/ (entities, typed relations, provenance, profiles, a journal). It is the shared world model that survives context windows. Treat it as memory you read at the start of work and write to as you develop. Zero-config commands.

codician-team/growmos · 606 tokens

ccjk CLAUDE.md

Instructions for miounet11/ccjk: 持久知识唯一来源：docs/CLAUDE-NOTES.md（请优先阅读本文档）.

miounet11/ccjk · 135 tokens

shinobi CLAUDE.md

Claude Code instructions for numbererikson/shinobi, covering claude.md, what this is, dogfooding — use shinobi to build shinobi, dev commands and layout.

numbererikson/shinobi · 1,118 tokens

skill-memory-bank CLAUDE.md