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.
npx skills add Ozzeron/prompt-pack --skill token-disciplinegit clone --depth 1 https://github.com/Ozzeron/prompt-packWrote 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/skills/ozzeron/prompt-pack/token-discipline)<a href="https://agentmods.dev/skills/ozzeron/prompt-pack/token-discipline"><img src="https://agentmods.dev/badge/skills/ozzeron/prompt-pack/token-discipline/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/skills/ozzeron/prompt-pack/token-discipline"><img src="https://agentmods.dev/badge/skills/ozzeron/prompt-pack/token-discipline.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 | $0.00087 | $0.01391 |
| Opus 5 | $0.00044 | $0.00696 |
| Sonnet 5 | $0.00017 | $0.00278 |
| Haiku 4.5 | $0.00009 | $0.00139 |
Grade A, and why
token-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 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.
How it starts
The opening of the file, as written. The whole thing — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Token Discipline
Tokens are an attention budget, not a money budget. The reason this skill exists is not to save the user money — even on unlimited budgets, a context window stuffed with low-signal files makes the model worse: it loses track of what matters, mixes patterns from unrelated code, and hallucinates more confidently. Bigger context is not smarter context.
The goal of this skill is not to make the agent read as little as possible. It is to make the agent read the right things first, and only widen the read when correctness demands it. Spend context where it buys reliability; do not spend it on noise.
This document is the single source of truth for "what an agent should and should not read" while working on a task. Every other prompt in this pack inherits these rules. Specific roles may add stricter rules on top, but never relax these.
When to use
Always. This is a cross-cutting concern, not a triggered behaviour.
Hard rules — never read
These directories and file types are noise by default. Do not read them raw under any circumstance unless the user explicitly asks for it.
- Dependency caches:
node_modules/,vendor/,.venv/,venv/,__pycache__/,.gradle/,.cargo/,target/,Pods/,DerivedData/ - Build outputs:
dist/,build/,out/,.next/,.nuxt/,.svelte-kit/,.turbo/,.parcel-cache/,coverage/ - VCS internals:
.git/,.hg/,.svn/ - Editor / OS:
.vscode/,.idea/,.DS_Store,Thumbs.db - Heavy binaries:
.exe,.dll,.so,.dylib,.bin,.iso,.zip,.tar,.7z,.dmg,.pdfover 1MB unless required - Heavy data:
.db,.sqlite,.parquet,.csvover 1MB,.jsonlover 1MB,.logover 200KB
If you need information from one of these, use a tool (grep, jq, sqlite query, scripts) instead of reading raw.
Soft rules — read with care
- Files larger than 50KB: state why you need them and ask before reading the full file.
Prefer
readwithoffset/limitorgrepto extract the relevant section. - Files larger than 200KB: must use offset/limit or a script-based extraction. Never read in full unless the user has explicitly asked.
- Generated artifacts (
*.lock,*.min.js,*.bundle.js): treat as binary, do not read. - Migration files: read only the most recent ones unless the task is migration history.
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.
- 3d ago Changed · +2 lines · +66 tokens per session 4b37cc3b5731
- 10d ago First seen · 117 lines · 21 tokens per session scan A 974a8ff6950a
token-discipline is a skill published in the GitHub repository Ozzeron/prompt-pack (8 stars, last pushed 4d ago), licensed MIT. It adds 87 tokens to every session and 1,391 once invoked, about $0.0004 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.
Other skills, from other repositories
gentle-ai-collab-perfect
Trigger: contributing to Gentleman-Programming/gentle-ai as an external collaborator. Strict issue-first workflow, honest PR bodies, contributor-vs-maintainer scope, chained-PR strategy, verification protocol, docstring coverage. Load whenever the active repo is Gentleman-Programming/gentle-ai and any part of the…
issue-creation
Trigger: issue creation, bug reports, feature requests, or issue approval. Create and triage GitHub issues from repository evidence.
sdd-tasks
Break an SDD change into implementation tasks. Trigger: orchestrator launches task planning for a change.
work-unit-commits
Plan commits as reviewable work units. Trigger: implementation, commit splitting, chained PRs, or keeping tests and docs with code.
systemic-issue-triage
Trigger: new issue, bug report, triage, backlog, issue flood, community report, root cause, dead-end, blocked user. Attack issues by root class, never one-by-one; fixes must shrink the system, not grow it.
sdd-research
Trigger: SDD research, external evidence, source-backed research. Produce auditable evidence for a selected research lane.