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 martinemde/dotfiles --skill prompt-engineeringgit clone --depth 1 https://github.com/martinemde/dotfilesWrote 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/martinemde/dotfiles/prompt-engineering)<a href="https://agentmods.dev/skills/martinemde/dotfiles/prompt-engineering"><img src="https://agentmods.dev/badge/skills/martinemde/dotfiles/prompt-engineering/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/martinemde/dotfiles/prompt-engineering"><img src="https://agentmods.dev/badge/skills/martinemde/dotfiles/prompt-engineering.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 90 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00000 | $0.00992 |
| Opus 5.5 | $0.00000 | $0.00397 |
| Sonnet 5.5 | $0.00000 | $0.00198 |
| Haiku 4.5 | $0.00000 | $0.00099 |
Grade A, and why
prompt-engineering 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 1mo 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Prompt Engineering
Prompt engineering is experimental design, not magic phrases. A good prompt is a contract: it states the job, separates trusted instructions from untrusted data, defines what success looks like, names the output shape, scaffolds genuine difficulty, and improves through evals rather than vibes.
Arguments
$ARGUMENTS
Reference
PLAYBOOK.md, alongside this file, holds the condensed principles, anti-patterns, task
patterns, and review rubric. Read it once per session before producing serious output.
Mode
Read the mode off the arguments: draft/write/create or a bare task description means draft
from scratch; review/audit or a pasted prompt with no other directive means review against
the rubric; improve/refine/fix with a prompt and its failure modes means revise it;
explain/why with a principle name means explain it. With no arguments, look for a
prompt-in-progress in the conversation and treat it as review — or say briefly what you can help
with if there isn't one.
Ambiguous signals resolve by inference, not by asking. Pick the mode that fits the inputs and note the inference in one line as you go. Ask only when proceeding would mean guessing a hard constraint — audience, schema, length cap — that changes the deliverable.
Gates
Every mode checks the prompt, existing or in progress, against these from PLAYBOOK.md:
- Goal clarity — outcome, audience, deliverable named
- Instruction/data separation — untrusted input fenced in tags or delimiters
- Output contract — schema, length, format measurable rather than adjectival
- Examples — present when format, taste, or classification boundaries matter; balanced and edge-case-aware
- Reasoning scaffold — matched to task shape (CoT, plan-and-solve, step-back, least-to-most, ReAct, PoT) rather than "show all your work" by default
- Grounding — for factual work: sources, citations, abstention rule, conflict handling
- Verification — a checklist against a rubric, source, or test, not "now check yourself"
- Long-context handling — instructions first, documents tagged, query at the end
- Failure mode — an explicit "if you cannot, say what is missing"
- Testability — could this be graded on an eval set?
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.
- 1mo ago First seen · 91 lines · 0 tokens per session scan A 7418bfe11e0d
prompt-engineering is a skill published in the GitHub repository martinemde/dotfiles (10 stars, last pushed 4d ago), licensed ISC. It costs nothing until one of its globs matches a file; then it loads 992 tokens. 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
system-prompts
Design AI app system prompts: instructions, context, tools, and output contracts.
prompt-design
Design production LLM or agent prompt stacks: instructions, tool contracts, examples, outputs, and runtime context. Use for precedence, conflicts, dynamic or untrusted context; prove behavior with agent-evaluation.
databricks
Operate Databricks bundles, jobs, pipelines, and Unity Catalog.
colab
Run code on Colab GPUs or TPUs and track compute units.
hf
Operate Hugging Face models, datasets, Spaces, and jobs with hf.
kaggle
Operate Kaggle competitions, datasets, kernels, and submissions.