oh-my-posh is a customizable prompt theme engine that renders command-line prompts across shells and platforms. Developers and terminal users use it to configure primary, secondary, right, and transient prompts in shells such as PowerShell, zsh, Bash, and cmd. The catalogue entries provide skills, instructions, agents, and an MCP integration for working with oh-my-posh.
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 JanDeDobbeleer/oh-my-posh --skill code-changesgit clone --depth 1 https://github.com/JanDeDobbeleer/oh-my-poshWrote 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/jandedobbeleer/oh-my-posh/code-changes)<a href="https://agentmods.dev/skills/jandedobbeleer/oh-my-posh/code-changes"><img src="https://agentmods.dev/badge/skills/jandedobbeleer/oh-my-posh/code-changes/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/jandedobbeleer/oh-my-posh/code-changes"><img src="https://agentmods.dev/badge/skills/jandedobbeleer/oh-my-posh/code-changes.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 | $0.00048 | $0.00987 |
| Opus 5.5 | $0.00019 | $0.00395 |
| Sonnet 5.5 | $0.00010 | $0.00197 |
| Haiku 4.5 | $0.00005 | $0.00099 |
Grade A, and why
code-changes 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 5d 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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code changes
1. Analyze, then stop
- Read the code, not only the issue, review comment, or stack trace: reports and bot reviewers are often wrong. Fetch
context with
gh issue view <n> --commentsorgh pr view <n> --comments. - Reproduce the problem when possible; say so when you cannot.
- Look for what already exists before designing anything: a helper, a similar module, a pattern another feature uses, a CLI to call instead of reimplementing it. Prefer the smallest change that reuses existing code, and the change that removes code over the one that adds it. Deprecated logic gets removed, not maintained.
- Report the root cause with file references, the proposed change and its scope, what stays out of scope, and open questions.
- Stop and wait for a go. Skip the stop only when the request already asks for the implementation ("fix it", "do it"). A go for analysis is not a go for implementation, and a new diagnosis after a failed verification needs a new go.
For triage only ("look at #n"), the report is the deliverable: add the blast radius and a suggested reply to the issue.
2. Plan
- Load the language and format skills for every file you will touch before writing any code, and follow them from the first line. A rule read at review time turns a violation into rework.
- When replacing working code, list what the old code does (defaults, key bindings, error messages) and keep all of it. Dropping any of it needs a go first.
- For a judgment call you cannot make with confidence (unclear root cause, security, irreversible change), ask the strongest available model that one question, then continue.
3. Implement
- Do standard work yourself. Delegate only independent tasks that can run in parallel (one worktree each) or mechanical batches for a cheaper model.
- A delegation carries the decided approach, the files, the binding skill rules, the verification commands, the non-goals, and the instruction to stop and report anything the spec does not cover.
- Never delegate analysis, final verification, commits, or pushes.
What ships with it
11 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.
- references/analyze.md 3.1 KB
- references/artifacts.md 5.3 KB
- references/delegate.md 2.4 KB
- references/deliver.md 1.4 KB
- references/escalate.md 2.4 KB
- references/issue-triage.md 1.7 KB
- references/model-tiers.md 3.8 KB
- references/plan.md 3.2 KB
- references/pr-review-comments.md 2.8 KB
- references/supervise.md 3.6 KB
- references/verify.md 4.0 KB
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.
- 5d ago Changed · -2 lines · -40 tokens per session 5979c78ddd66
- 1mo ago First seen · 81 lines · 88 tokens per session scan A 771a48bafca9
code-changes is a skill published in the GitHub repository JanDeDobbeleer/oh-my-posh (23,592 stars, last pushed today), licensed MIT. It adds 48 tokens to every session and 987 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
printing-press-output-review
Internal sub-skill: agentic review of a printed CLI's sampled command output for plausibility issues that rule-based checks can't encode (substring-match relevance, format bugs, silent source drops, ranking failures). Invoked via the Skill tool by the main printing-press skill at Phase 4.85 and printing-press-polish…
superpowers-zh
Use when constraining AI coding with Chinese TDD methodology, systematic debugging, code review, and verification workflows. Superpowers-zh: Chinese adaptation of the Superpowers AI-assisted programming skills and methodologies.
sillytavern-overseer
A strict review and debugging workflow built around monitoring progress, finding quality problems, applying pressure, and checking the final result against a standard.
theme-arena
A Chinese-language problem-solving framework that approaches debugging and reviews as a structured five-stage technical contest.
theme-sect-discipline
A themed code-quality and review guide built around inspection, finding problems, correction, and acceptance. It includes checks for reading failure details, inspecting source context, validating assumptions, and reproducing issues in isolation.
fix-llm-artifacts
Applies fixes from a prior review-llm-artifacts run, with safe/risky classification. Respects verify-llm-artifacts output when present to skip false positives.