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 agentmods add skills/zeroaurora/stdskill/standardnpx skills add ZeroAurora/stdskill --skill standardgit clone --depth 1 https://github.com/ZeroAurora/stdskillWhat 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 | $0.00030 | $0.01028 |
| Opus 5 | $0.00015 | $0.00514 |
| Sonnet 5 | $0.00006 | $0.00206 |
| Haiku 4.5 | $0.00003 | $0.00103 |
Grade B, and why
standard scanned grade B with 2 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 yesterday.
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.
Unrestricted tool accessmediumExcessive agency
A wildcard tool grant or "run any command" leaves no least-privilege boundary at all.
3. Agents MUST NOT continue any tasks, edit any code or execute any commands if they fail to comprehend the usage of any tools or libraries used in the project due to a lack of documentation and knowledge. Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
2. If no tool is available, Agents MUST NOT fetch documentation by themselves using raw HTTP requests. For example, agents must not use `curl` command to fetch websites. How it starts
The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Preface
This skill defines a common set of standard rules for coding agents ("agents" for short) to follow. Bad things will happen if agents don't obey these rules.
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this skill are to be interpreted as described in RFC 2119.
1. General Rules
- Agents MUST comply with users' requirements. They MUST NOT perform any steps, operations, or actions beyond the scope of those requirements.
- Agents MUST NOT guess or assume anything that is not explicitly present in the context. When available, agents SHOULD use tools to ask users clarifying questions.
2. Tool Utilization Rules
- Agents MUST make full use of all tools provided to them.
- When MCP tools are available, agents SHOULD prioritize them over other tools.
- For example, if
mcp__filesystem__list_directorytool is available, agents should use it to read a directory instead of executinglscommand.
bashtool or any other tools that execute a command SHOULD be used only if the required feature does not exist in any other tools OR if all other tools fail to accomplish the goal.
- For example, if there are
edittool available,bashtool withcatcommand should not be used.
3. Code Editing Rules
- Agents SHOULD NOT edit any files or directories related to coding standards, including but not limited to linter configuration files, formatter configuration files, and any other files that are used to configure the coding environment. If an edit is absolutely required, Agents MUST ask users for confirmation.
- Agents MUST NOT edit any files or directories that are not part of the project. The only exception is that Agents MAY edit files or directories inside
/tmp,%TEMP%, or any other temporary directories specified by users.
4. Code Analysis Rules
- Agents SHOULD frequently check LSP (Language Server Protocol) messages (if available), linting outputs and type checking results. An execution of linter or type checking tool SHOULD be triggered after any complete edit action to the code.
- Agents SHOULD do everything possible to fix any errors or warnings by reviewing the code in detail and editing them. If it is not possible, Agents MUST stop any current tasks and ask users for help.
- Agents MUST NOT try to bypass any errors or warnings raised by compilers, linters or static code analyzers. Forbidden methods include but not limited to:
- Disabling or suppressing compiler warnings or linter rules by inserting disabling comments or editing configuration files.
- Using compiler flags that disable type checking or other features that are intended to catch errors.
- Evading typing enforcement by using type casting (for example,
as unknown asin TypeScript) or other methods that bypass type checking. - Downgrading, replacing or modifying third-party dependencies that are required for the project to compile or run, without explicit permission from users.
- Any other methods that are intended to bypass enforcements of coding standards.
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.
- yesterday First seen · 60 lines · 30 tokens per session scan B 9ca3ab3ea084
standard is a skill published in the GitHub repository ZeroAurora/stdskill (9 stars, last pushed 5mo ago), licensed CC0-1.0. It adds 30 tokens to every session and 1,028 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 2 findings (unrestricted tool access, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
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…