documentation-budget

A classification skill for deciding whether an AI-facing document is needed, how detailed it should be, and how to keep it small.

In plain words
What is it for?
Use it before creating or revising specifications, decisions, task lists, verification gates, handoffs, backlogs, problem records, or product and business-logic documents.
Why use it?
It limits persistent documentation to cases where it saves more future context and coordination effort than it costs to maintain and reload.

Skill for Claude CodeCodex

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 skills/fusebase-dev/fusebase-flow/documentation-budget
Any agent
npx skills add fusebase-dev/fusebase-flow --skill documentation-budget
Clone the repo
git clone --depth 1 https://github.com/fusebase-dev/fusebase-flow

Made for: Claude Code, Codex.

Per session 123 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,873 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 $0.00123 $0.02873
Opus 5 $0.00062 $0.01437
Sonnet 5 $0.00025 $0.00575
Haiku 4.5 $0.00012 $0.00287

Measured 2d ago against content hash ac49c0e5f603, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

documentation-budget 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.

.agents/skills/documentation-budget/SKILL.md · 185 lines

How it starts

The opening of the file, as written. The whole thing — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Documentation Budget

Style: Mode-B-lite. Pre-write classifier for AI-consumed artifacts. Operationalizes FR-23. Reference this skill from doc-producing skills; do not duplicate it.

Purpose

Make documentation proportional to risk, complexity, and future AI value. An AI-consumed artifact is created only when it reduces future context cost more than it adds. Duplicate rationale, narrative padding, and unnecessary persistent docs are paid in tokens on every future load and spawn stale conflicting copies. This is the documentation-axis complement to FR-21 (lightweight-lane, which scales process ceremony): FR-21 decides how much process a change carries; FR-23 decides how much persistent documentation it leaves behind.

When to invoke

Before creating, expanding, or revising any of:

  • docs/specs/<slug>/spec.md, decisions.md, tasks.md, verification-gate.md
  • docs/tmp/handoff.md (active session continuity)
  • docs/tmp/handoff/<YYYY-MM-DD>-<slug>-{implement,deploy}.md (formal role-relay)
  • docs/changes/<date>-<slug>.md (Lightweight change-note)
  • docs/backlog/**, docs/problem-catalog/**
  • docs/<app>/product.md, docs/<app>/business-logic.md, docs/<app>/business-logic-index.md
  • project-internal skills under docs/skills/**
  • any persistent AI-consumed doc not in the carve-out below

Do not invoke when

  • Writing operator chat only (Mode A).
  • Writing or editing source comments — use comment-policy (FR-22).
  • Editing human-facing files only: README.md, LICENSE, PUBLISHING.md, CONTRIBUTING.md, AGENTS.md, CLAUDE.md, GEMINI.md — unless the edit changes AI workflow behavior.
  • No persistent artifact will be created or changed.

Required inputs

Input Where it lives If missing
The artifact request operator intent / current ticket Ask what future action the doc must enable
Change risk + size spec / lane classification (lightweight-lane, FR-21) Default to higher tier (fail-safe-up)
Existing artifacts docs/specs/<slug>/, docs/<app>/, git history Read before writing; prefer a pointer to an existing owner

Read the full file on GitHub · 185 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 · 185 lines · 123 tokens per session scan A ac49c0e5f603

Subscribe to this mod's changes

documentation-budget is a skill published in the GitHub repository fusebase-dev/fusebase-flow (9 stars, last pushed 7d ago), licensed MIT. It adds 123 tokens to every session and 2,873 once invoked, about $0.0006 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 2026-08-31.

Related

Other skills, from other repositories

clickup-cli

ClickUp CLI for managing tasks, sprints, comments, statuses, and Docs. Use when the user needs to interact with ClickUp — creating/editing tasks, checking sprint status, adding comments, linking PRs, managing Docs and pages, or searching tasks. Prefer this CLI over raw API calls.

ulises-jeremias/agent-toolkit · 65 tokens

test-estimation

Estimate testing effort, duration, and resourcing for a Katalon True Platform/TestOps cycle. Use when the question is how long testing will take, how many testers it needs, whether the scope fits the sprint window, or what a scope change costs in person-hours. Sizes design, manual execution, automated execution and…

katalon-labs/true-skills · 198 tokens

test-reporting

Report Katalon True Platform/TestOps quality metrics to people outside QA. Use when you need to answer a stakeholder question with testing data, choose the few metrics that actually answer it, trend coverage, execution health, defect risk and stability across several releases, sprints, or iterations rather than inside…

katalon-labs/true-skills · 181 tokens

release-analyze

Analyze Katalon True Platform/TestOps release readiness from testing quality data. Use when you need to use Katalon MCP metrics and results to assess whether a release, sprint, iteration, version, test plan, suite, or repository is ready to ship; summarize requirement coverage, execution health, defect risk, test…

katalon-labs/true-skills · 111 tokens

test-plan

Plan Katalon True Platform/TestOps testing for a release, sprint, or feature. Use when you need to translate quality goals into scope, prioritize testing by requirement coverage and risk, decide what to test first, or build the executable plan structure (folders, suites, and sprint/release association) that stands in…

katalon-labs/true-skills · 140 tokens

uv-usage

Provides concise guidance for using uv (Python package manager), including project workflows, pip-compatible commands, Python version management, and PEP 723 inline script dependencies. Use when users mention uv, uv run, inline dependencies, PEP 723, or Python dependency/project management.

liatrio-labs/ai-prompts · 59 tokens