fill

A command that loads a project's Memory Bank, a set of documents containing its product details, requirements, design, and working rules, into the agent's context.

In plain words
What is it for?
Use it at the start of a task when the project stores its context in a .memory-bank folder.
Why use it?
It gives the agent the project background before work begins and identifies missing information or useful documents to read next.

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/nicelight/devrails-26/fill
Clone the repo
git clone --depth 1 https://github.com/nicelight/DevRails-26
Per session 20 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 443 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 $0.00020 $0.00443
Opus 5 $0.00010 $0.00221
Sonnet 5 $0.00004 $0.00089
Haiku 4.5 $0.00002 $0.00044

Measured 3d ago against content hash 721d7f77f9e8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

fill 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 3d 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.

skills/_shared/references/commands/fill.md · 44 lines

What it actually says

/fill — Prime context

  1. Прочитай .memory-bank/index.md.
  2. Прочитай .memory-bank/mbb/index.md.
  3. По аннотированным ссылкам открой релевантные документы:
    • product.md, requirements.md, нужные features/FT-XXX, architecture/*, guides/*.
    • если присутствуют и релевантны теме: spec-backbone.md, spec-index.md, invariants.md, glossary.md, contracts/*, states/*, runbooks/*, testing/*.
  4. Если контекста не хватает, верни gaps и рекомендуемые следующие reads. Используй direct reads по умолчанию. /context-manifest можно только рекомендовать для действительно широкого discovery, когда оно дешевле direct reads и delegation разрешено текущей ролью/operator contract.
  5. Перед началом работы сформулируй коротко:
    • что прочитано
    • gaps и рекомендуемые следующие reads
    • цель
    • критерии done
    • применимые quality gates

Если пользователь передал $ARGUMENTS, используй их как фокус:

  • «Запраймься по теме: $ARGUMENTS»

<hard_invariants>

  • /fill строго read-only во всех ролях и режимах.
  • Не создавай и не обновляй .protocols/*, task plan или другой artifact.
  • Не запускай сабагентов, delegation или /context-manifest из /fill.
  • Если нужен protocol mutation, назови существующий owning execution/planning skill как следующий шаг, но не выполняй mutation. </hard_invariants>
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. 3d ago First seen · 44 lines · 20 tokens per session scan A 721d7f77f9e8

Subscribe to this mod's changes

fill is a command published in the GitHub repository nicelight/DevRails-26 (22 stars, last pushed 5d ago), licensed MIT. It adds 20 tokens to every session and 443 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.