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/brief)<a href="https://agentmods.dev/commands/fockus/skill-memory-bank/brief"><img src="https://agentmods.dev/badge/commands/fockus/skill-memory-bank/brief/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/brief"><img src="https://agentmods.dev/badge/commands/fockus/skill-memory-bank/brief.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 | $0.00043 | $0.01594 |
| Opus 5.5 | $0.00017 | $0.00638 |
| Sonnet 5.5 | $0.00009 | $0.00319 |
| Haiku 4.5 | $0.00004 | $0.00159 |
Grade A, and why
brief 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 14d 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 — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/mb brief [--request | --request-file ] [--input ]… [--auto]
Formalize a raw request into a one-page brief stored with its sources, so the work
enters the pipeline already shaped: brief → discuss → sdd → work.
This command is deliberately NOT an interview. /mb discuss owns the deep
elicitation; briefing asks at most a handful of questions and only when it has to.
When to use
At the very start, when the request is a paragraph of prose plus a pile of
documents and nobody has written down what the work actually is. Skip it when the
request is already one clear sentence — go straight to /mb discuss.
Arguments
<topic>— matches[a-z0-9][a-z0-9-]*. Anything else iserror=usage, exit 2, before anything is written.--request <text>— the request inline.--request-file <path>— the request read from a file.--input <path>— repeatable; a source document to scan and copy intoinputs/.--auto— skip the clarifying-question gate and infer on best effort.
There is no overwrite or re-publish flag in MVP. Re-formalizing is done on a new
<topic>, or after the user deletes briefs/<topic>/ by hand.
Pre-flight
Resolve the skill bundle root once and invoke every bundled helper through it — the working directory during /mb brief is the user's project, so a bare relative scripts/mb-… call either exits 127 or runs somebody else's script:
SKILL_DIR="${MB_SKILLS_ROOT:-$HOME/.claude/skills/memory-bank}" # memory-bank skill bundle root
[ -f "$SKILL_DIR/scripts/mb-brief.sh" ] || {
echo "mb: skill bundle not found at $SKILL_DIR — set MB_SKILLS_ROOT" >&2; exit 2; }
Never derive the root from $0: in an executable-Markdown snippet $0 is the shell, not this file.
Request source
Exactly one source of request text is used, and the rule is enforced before any candidate file is written.
- Passing both
--requestand--request-fileiserror=usage, exit 2; when neither flag is given, the request is only the text of the user message that carried the invocation. - A
--requestwhose value has no non-whitespace character iserror=request_emptybefore the candidate is written, exit 2. - A
--request-filethat is missing, a symlink, a directory or contains..iserror=request_unreadable path=<p>before the candidate is written, exit 2. - A
--request-filewhose contents have no non-whitespace character iserror=request_empty path=<p>, likewise before the candidate is written.
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.
- 14d ago Changed · +25 tokens per session f327a9470529
- 1mo ago First seen · 113 lines · 18 tokens per session scan A e07266fbb625
brief is a command published in the GitHub repository fockus/skill-memory-bank (29 stars, last pushed 3d ago), licensed MIT. It adds 43 tokens to every session and 1,594 once invoked, about $0.0002 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 2026-08-30.
Other commands, from other repositories
edit
Edit an existing BPMN diagram, then re-layout, validate and lint.
documenter
Documentation with batch file operations for comprehensive docs.
research-verify
Verify existing research findings against independent primary sources. Upgrades confidence from 'sources agree' to 'independently verified.'.
spreadsheet-audit
Perform a 6-step audit of an Excel spreadsheet for formula correctness, formatting consistency, data integrity, boss-auditable simplicity, and spec compliance. Uses the excel-screenshot MCP server tools — no Excel installation required.
audit
Audit closed tickets for regressions against today's code (unchanged, evolved, implemented wrong, not implemented), plus duplicate and groupable open ones.
wiki-tune
Tune wiki schema and conventions interactively.