discipline

discipline is a skill for Claude Code, Codex from fledgeling-co/fledgeling-plugins. It costs 140 tokens per session (7,557 once invoked), scanned A, original, MIT.

A set of session-start rules for reducing repeated input when using Claude Opus 5 for coding tasks. It includes a fixed instruction block and evidence files explaining the choices behind it.

In plain words
What is it for?
Use it when tuning coding-agent sessions for lower token use, reviewing the evidence behind those settings, or checking changes before committing them.
Why use it?
It addresses token waste caused by repeatedly sending, rereading, or reopening the same context. The supporting checks help detect changes to the block or unsupported figures.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions CLAUDE.md; mentions subagents; mentions Claude Code.

Part of the discipline plugin — 1 skill shipped together

Good fit Use it when tuning coding-agent sessions for lower token use, reviewing the…

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/fledgeling-co/fledgeling-plugins/discipline
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.

Any agent
npx skills add fledgeling-co/fledgeling-plugins --skill discipline
Clone the repo
git clone --depth 1 https://github.com/fledgeling-co/fledgeling-plugins

Made for: Claude Code, Codex.

Or install discipline, the plugin that ships this one along with the rest of its 1 skill.

Wrote 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.

agentmods badge for discipline

README.md
[![agentmods](https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/discipline.svg)](https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/discipline)
Your own site
<a href="https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/discipline"><img src="https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/discipline.svg" alt="Measured on agentmods" height="20"></a>
Per session 140 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 7,557 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00140 $0.07557
Opus 5 $0.00070 $0.03778
Sonnet 5 $0.00028 $0.01511
Haiku 4.5 $0.00014 $0.00756

Measured 6d ago against content hash 29d8f484bb33, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

discipline 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 6d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/block-check.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

plugins/discipline/skills/discipline/SKILL.md · 456 lines

How it starts

The opening of the file, as written. The whole thing — 456 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Token discipline

Most of what a session spends is not what anyone typed. Every turn resends the conversation from the top, and cache reads bill at 0.1× base input — so the lever is not shorter prose. It is not resending, not re-reading, and not re-opening a fresh prefix somewhere else.

This skill is two artifacts, and keeping them apart is the point:

Artifact What it is Where it lives
The block 881 bytes (~220 tokens, v4) injected into the request's system field at session start references/injected-block.md — verbatim, byte-stable
This file Why the block says what it says, what was rejected, what is measured vs assumed Read when editing the block or debugging a session

Two supporting files carry what this one asserts. references/evidence.md holds every measurement in full; references/provenance.md is the registry that marks each figure on two closed axes, and scripts/block-check.py reads that registry and fails the build on an untiered figure, a drifted literal, or an assumed number stated in the argument. Run it before any commit that touches the block or a number:

python3 skills/discipline/scripts/block-check.py --verbose   # must exit 0

The block is not this file compressed. It is a separate, versioned, size-pinned literal.

v1 (1,029 bytes) and v3 (736 bytes) still exist in the code and must stay there. PERCH-0327 pins the block per conversation and replays exact bytes, so a conversation opened before v4 keeps emitting its own version for the rest of its life. Deleting or tidying either re-mints every older conversation at full price.

Running as a Gemini model? Read gemini.md in this directory first, then follow this file with the overrides it names. It routes the block-editing lane by measured shape, fills the bound ledger from block-check.py's own output rather than from this file's prose, and names clause 6 as the clause this model family actually needs. Other models skip it.

Read the full file on GitHub · 456 lines

Files

What ships with it

6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 6d ago First seen · 456 lines · 140 tokens per session scan A 29d8f484bb33

Subscribe to this mod's changes

discipline is a skill published in the GitHub repository fledgeling-co/fledgeling-plugins (2 stars, last pushed today), licensed MIT. It adds 140 tokens to every session and 7,557 once invoked, about $0.0007 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.