flow

A command for manually choosing a work route, such as bug fix, code change, architecture, migration, or research.

In plain words
What is it for?
Use it to select a route, preview the result with a dry run, and let the resolver apply any required minimum level of work.
Why use it?
It lets you bypass automatic task classification when you already know which route the work needs.

Command

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 commands/fockus/skill-memory-bank/flow
Clone the repo
git clone --depth 1 https://github.com/fockus/skill-memory-bank
Per session 15 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 691 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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.

ModelPer sessionOnce invoked
Fable 5 $0.00015 $0.00691
Opus 5 $0.00008 $0.00345
Sonnet 5 $0.00003 $0.00138
Haiku 4.5 $0.00002 $0.00069

Measured 2d ago against content hash 4c857221a439, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

flow 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 2d 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.

commands/flow.md · 67 lines

How it starts

The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/mb flow

Explicitly select the flow route, skipping auto-classification. This is the escape-hatch (REQ-DF-025), mirroring the /mb work --workflow <name> precedent: auto-routing via /mb analyze-task is the default, and /mb flow is the manual override when you already know the route.

/mb flow <route>          # e.g. /mb flow arch
/mb flow --route <route>  # equivalent flag form

<route> is one of: bugfix | code-change | arch | migration | research.

What it does

  1. Resolve the active Memory Bank (scripts/_lib.sh::mb_resolve_path).

  2. Skip classification — there is no LLM step. Hand the named route straight to the resolver, which still applies the deterministic floor + writes the fence:

    SKILL_DIR="$(cd "$(dirname "$0")/.." && pwd)"   # memory-bank skill bundle root
    bash "$SKILL_DIR/scripts/mb-flow-route.sh" --route <route>
    
  3. Report the resolved route from the resolver's JSON. If floor_triggered is true, the override was raised to the floor — say so and name the reasons.

Add --dry-run to preview without writing the fence. The resolver derives the changed-file scope from git on its own; pass --changed/--changed-file only to drive it explicitly.

The override does NOT bypass safety

Choosing a route here changes only the starting candidate — it does not disable any guard:

  • Route-floor still applies (REQ-DF-022). An override below the floor is raised to the floor, never honored blindly (REQ-DF-025). /mb flow bugfix on a diff that touches domain/, an application/ports path, a *Protocol/ABC/contract file, a declared protected_path, or a linked plan with depends_on > 0 resolves to arch, not bugfix.
  • Firewall still applies (REQ-DF-040). The done-gate is still scripts/mb-flow-verify.sh; the flow is never declared finished on red.

So the override can only raise a route or pick among routes at/above the floor — it can never route under the deterministic floor.

Read the full file on GitHub · 67 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.

  1. 2d ago First seen · 67 lines · 15 tokens per session scan A 4c857221a439

Subscribe to this mod's changes

flow is a command published in the GitHub repository fockus/skill-memory-bank (25 stars, last pushed 1mo ago), licensed MIT. It adds 15 tokens to every session and 691 once invoked, about $0.0001 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 2026-08-30.