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 commands/fockus/skill-memory-bank/startgit clone --depth 1 https://github.com/fockus/skill-memory-bankWrote 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/commands/fockus/skill-memory-bank/start)<a href="https://agentmods.dev/commands/fockus/skill-memory-bank/start"><img src="https://agentmods.dev/badge/commands/fockus/skill-memory-bank/start.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 | $0.00007 | $0.01099 |
| Opus 5 | $0.00003 | $0.00549 |
| Sonnet 5 | $0.00001 | $0.00220 |
| Haiku 4.5 | $0.00001 | $0.00110 |
Grade A, and why
start 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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Canonical session-start command. /mb start is an alias that dispatches here.
Pre-flight: resolve Memory Bank path
Memory Bank may live in one of three places. Always resolve the active bank through mb_resolve_path (in scripts/_lib.sh) — never assume ./.memory-bank/ is the only signal:
- Local —
<project>/.memory-bank/(team-shared, default of/mb init). - Global —
<agent_config>/memory-bank/projects/<id>/.memory-bank/registered in<agent_config>/memory-bank/registry.json(personal storage chosen via/mb init --storage=global --agent=<name>). - Legacy —
.claude-workspaceexternal pointer (still supported for backward compatibility).
When none of the above resolves, surface [MEMORY BANK: ABSENT] and stop the lifecycle flow. Do not auto-initialize — the user may be in rules-only mode intentionally (TDD/SOLID/Clean Architecture/DRY/KISS/YAGNI/Testing Trophy still apply to code work; only Memory Bank commands stay inactive).
Pre-flight: detect v1 layout
Before loading context, check whether the project is still on v1 Memory Bank naming. Run:
ls .memory-bank/ 2>/dev/null | grep -E '^(STATUS|BACKLOG|RESEARCH|plan)\.md$'
If any of STATUS.md / BACKLOG.md / RESEARCH.md / plan.md appear (and the corresponding lowercase variant — status.md / backlog.md / research.md / roadmap.md — is NOT present), the project is on v1 and must be migrated.
Tell the user exactly:
"Detected v1 Memory Bank layout (uppercase STATUS.md / BACKLOG.md / plan.md). v2 requires lowercase names. Run:
bash ~/.claude/skills/memory-bank/scripts/mb-migrate-v2.sh --dry-runto preview, then
--applyto execute. Backup is created automatically."
Do NOT proceed with context loading until migration is done, UNLESS the user explicitly says "read v1 anyway" — in which case fall back to reading the old names.
1. Check whether Memory Bank is active
[ -d ./.memory-bank ] && echo "[MEMORY BANK: ACTIVE]" || echo "[MEMORY BANK: INACTIVE]"
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 · 96 lines · 7 tokens per session scan A 4cb9fe460560
start is a command published in the GitHub repository fockus/skill-memory-bank (25 stars, last pushed 1mo ago), licensed MIT. It adds 7 tokens to every session and 1,099 once invoked, about $0.0000 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 2026-09-03.
Other commands, from other repositories
catchup
Rebuild context after /clear or starting a new session.
notebook
Low-ceremony capture for ideas (half-formed, want to mature into stories), todos (concrete actions), and notes (durable project/team facts for future recall). One-line capture, conditional elaboration AUQ. Use BEFORE thoughts get forced into stories.
vitaecontext-vitaegraph
Create, validate, index, or use a private VitaeGraph.
Update
Update the Evyasys plugin and manage context compression — shows version diff, changelog highlights, asks for confirmation, then always asks about compression state. Project config, credentials, docs, and board artefacts are never touched.
memory-review
Run the memory review — staleness, duplicates, graduation candidates.
memory
Manage Grok cross-session memory.