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.
git 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/rules)<a href="https://agentmods.dev/commands/fockus/skill-memory-bank/rules"><img src="https://agentmods.dev/badge/commands/fockus/skill-memory-bank/rules/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/fockus/skill-memory-bank/rules"><img src="https://agentmods.dev/badge/commands/fockus/skill-memory-bank/rules.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00078 | $0.01729 |
| Opus 5.5 | $0.00031 | $0.00692 |
| Sonnet 5.5 | $0.00016 | $0.00346 |
| Haiku 4.5 | $0.00008 | $0.00173 |
Grade A, and why
rules 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 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.
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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/mb rules — Key rules selection: $ARGUMENTS
The ## Key rules block (<!-- mb-key-rules:start/end -->) sits at the top of CLAUDE.md and
AGENTS.md: one line per rule from the catalog rules/key-rules.json plus the user's own rules,
ending with a pointer to the detailed RULES.md. The selection lives in the key_rules field of the
rules profile (references/rules-profile.schema.md); project overrides user. The architecture and
test lines (TDD, Testing Trophy, coverage) and the four principles render from the profile's quality
settings — architecture and quality — changed with set; that is their only place.
With --scope=project the same settings are also written, human-readable, into the project
RULES.md (<repo>/RULES.md, else <bank>/RULES.md) between <!-- mb-project-rules:start --> and
<!-- mb-project-rules:end -->: architecture(s) with the preset rules and severity, principles,
TDD, Trophy, coverage thresholds. Text outside the block is the user's and never changes; only
init --scope=project creates a missing RULES.md.
The only writer is scripts/mb-rules.sh — never edit the block or the profile by hand:
bash "${MB_SKILLS_ROOT:-$HOME/.claude/skills/memory-bank}/scripts/mb-rules.sh" <subcommand> [args] [--scope=user|project]
Subcommands
| Subcommand | Effect |
|---|---|
list |
Numbered checklist by group: [x] on, [ ] off, [*] locked (always on); own rules numbered below |
enable <id> / disable <id> |
Turn a catalog rule on/off; disabling a locked rule → exit 2 with the reason. For solid/dry/kiss/yagni, tdd, testing-trophy, coverage this flips the quality setting |
add "<text>" |
Add an own rule (one line, ≤200 chars) |
remove <n> |
Remove own rule number n from list |
init [--enable=id,id] [--disable=id,id] [--custom="<text>"]... |
Replace the scope's whole selection (used by onboarding below) |
set <key> <value> |
Change one quality setting (table below) |
sync |
Re-render the block (and the project RULES.md block) from the profiles; --scope=project also refreshes the AGENTS.md Memory Bank block and existing per-host rule files — the command the session-start hint names when a block's mb-stamp is stale (MB_AUTO_REFRESH=on runs it automatically) |
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 · 86 lines · 78 tokens per session scan A 7e2dc86d2198
rules is a command published in the GitHub repository fockus/skill-memory-bank (26 stars, last pushed 3d ago), licensed MIT. It adds 78 tokens to every session and 1,729 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-08.
Other commands, from other repositories
implement
Execute tasks from a track's implementation plan following TDD workflow.
execute
Execute the next available tracked task with TDD, pre-commit validation, PR creation, AI code review, and issue tracker bridge sync.
tdd
Start test-driven development workflow for a feature or fix.
gsd:validate-phase
Retroactively audit and fill Nyquist validation gaps for a completed phase.
config
Complete configuration guide for pair programming sessions.
start
Start a new pair programming session with AI assistance.