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 OutlineDriven/odin-claude-plugin --skill punishing-practicesgit clone --depth 1 https://github.com/OutlineDriven/odin-claude-pluginWrote 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/outlinedriven/odin-claude-plugin/punishing-practices)<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/punishing-practices"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/punishing-practices/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/outlinedriven/odin-claude-plugin/punishing-practices"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/punishing-practices.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00046 | $0.00987 |
| Opus 5 | $0.00023 | $0.00494 |
| Sonnet 5 | $0.00009 | $0.00197 |
| Haiku 4.5 | $0.00005 | $0.00099 |
Grade A, and why
punishing-practices 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 — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Punishing practices
Contract
| Field | Bound contract |
|---|---|
| Trigger | A workflow, plan, diff, or completed work cycle must be checked for practices that punish the project later. |
| Authority | Read-only. No file, VCS, credential, paid, published, deployed, or remote mutation. |
| Side effect | Chat output naming each detected practice with evidence and a cheaper alternative, or a clean verdict. |
| Done | Every detected practice is named with an evidence line and a cheaper alternative (or explicit absence), or the artifact is reported clean with scan coverage stated. The report changes nothing. |
Refusals
- Remote, credential, publish, deploy, or irreversible changes: rejected. This skill emits a chat report only.
- Invented failures: rejected. If no practice class matches, report the artifact clean.
- Source or remote-system changes: rejected. This skill never edits code, branches, credentials, or remote state.
Inputs
- Artifact to audit (required): a workflow, plan, diff, or completed work cycle, supplied as a concrete artifact (file path, pasted content, committed diff range, workflow definition, or verbal cycle description).
- Success metric or budget (optional): the metric or budget the artifact is meant to serve, used only to judge whether a practice is score chasing or budget burn.
- Scope hint (optional): the user may bound the audit to a specific subsystem, time window, or decision.
Procedure
- Read the supplied artifact without mutating it. Bound the audit to the artifact's stated scope; do not widen to unreferenced code or history. Done when: the artifact is read and the audit scope is bounded.
- For each of the five practice classes, scan the artifact for a match and collect the exact evidence line (quote, step number, or configuration value) that proves it:
- Symptom patching: a change suppresses a signal (warning, error, failing test, metric) without addressing its cause.
- Infinite retries: a loop, hook, or agent step retries without a bounded attempt count, backoff ceiling, or stop condition.
- Weak verification: a check, gate, or test that can pass while the defect it claims to catch is still present (tautological assertion, mocked-out assertion, missing assertion, green-on-broken).
- Score chasing: work optimizes a metric or rubric score while the underlying outcome it proxies degrades or stays unknown.
- Budget burn: spend, tokens, time, or iterations accumulate without a ceiling or a stopping rule tied to the outcome. Done when: all five classes have been scanned and evidence collected or confirmed absent.
- For every match, name the practice class, quote the evidence line, and state one cheaper alternative that achieves the same goal without the punishing effect. If no cheaper alternative exists for a match, state that explicitly rather than inventing one. Done when: every match has its class, evidence, and alternative (or explicit absence) recorded.
- If no practice class matches, report the artifact clean with the scan coverage stated: name the five classes scanned and the artifact scope covered. Done when: the clean verdict is emitted.
- Emit the audit report as chat output. Do not edit, stage, commit, or open issues. Done when: the report is emitted as chat output and no mutation occurred.
What ships with it
1 file 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.
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 First seen · 53 lines · 46 tokens per session scan A 9a6c96db9559
punishing-practices is a skill published in the GitHub repository OutlineDriven/odin-claude-plugin (35 stars, last pushed 2d ago), licensed Apache-2.0. It adds 46 tokens to every session and 987 once invoked, about $0.0002 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 2026-09-06.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…