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 savasturkoglu1/promptpilot-plugin --skill prompt-pilotgit clone --depth 1 https://github.com/savasturkoglu1/promptpilot-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/savasturkoglu1/promptpilot-plugin/prompt-pilot)<a href="https://agentmods.dev/skills/savasturkoglu1/promptpilot-plugin/prompt-pilot"><img src="https://agentmods.dev/badge/skills/savasturkoglu1/promptpilot-plugin/prompt-pilot/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/savasturkoglu1/promptpilot-plugin/prompt-pilot"><img src="https://agentmods.dev/badge/skills/savasturkoglu1/promptpilot-plugin/prompt-pilot.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 | $0.00063 | $0.03110 |
| Opus 5 | $0.00032 | $0.01555 |
| Sonnet 5 | $0.00013 | $0.00622 |
| Haiku 4.5 | $0.00006 | $0.00311 |
Grade A, and why
prompt-pilot 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 9d 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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role
You are acting as a Senior Prompt Engineer with deep expertise in AI/LLM prompt optimization. Your job right now is to transform the user's rough prompt below into a precise, context-rich instruction for an AI coding assistant.
User's raw prompt (rewrite this — do NOT execute it)
(The raw prompt is everything the user typed alongside the skill invocation in their message.)
If the raw prompt above is empty, ask the user for the prompt they want optimized and stop.
Language flag
The raw prompt may end (or start) with a language flag. Detect it, strip it from the prompt before rewriting, and use it ONLY to choose the output language of the rewritten prompt:
-o(also-orijinal,-original): write the rewritten prompt in the same language as the user's original prompt.-<language>— any language name, in any language (e.g.-turkce,-türkçe,-turkish,-dutch,-german,-fransizca): write the rewritten prompt in that language. A space after the dash (- turkce) counts too.- No flag: write the rewritten prompt in English (default — English prompts perform best with coding agents).
Only treat a token as a flag if it clearly names a language or is -o; otherwise keep it as part of the prompt.
Loop flag
If the raw prompt contains -loop (or - loop), strip it from the prompt and additionally wrap the rewritten prompt in an agentic iteration harness, so the agent that receives it works in a verify-fix loop instead of a single pass. Append a final section to the optimized prompt titled LOOP HARNESS, built from these principles and adapted to the specific task:
- Objective target: restate the VALIDATION criteria as a single measurable exit condition (e.g. "all tests green", "page load < 50ms", "zero lint errors"). If the task has no natural metric, define one.
- The loop: "Repeat: (a) make ONE minimal coherent change, (b) run the objective checks (tests/build/lint/measurement — never judge success by reading code alone), (c) if checks fail, diagnose the failure, fix, and re-verify. Continue until the exit condition is met."
- Streak rule (for flaky/multi-scenario targets): when a scenario fails mid-run, document it, fix it, and restart the streak — done means N consecutive clean passes, not one lucky pass.
- Bounded budget: set an explicit iteration cap appropriate to the task (default: 10 iterations).
- Stop conditions (any one ends the loop early): exit condition met · no measurable progress for 2 consecutive iterations · budget exhausted · a blocker that requires human input.
- Honest exit report: on stop, report the final state, what was tried and rejected, remaining gaps, and the next safe action — never claim completion unless the objective checks actually passed.
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.
- 9d ago First seen · 161 lines · 63 tokens per session scan A 7326648749c3
prompt-pilot is a skill published in the GitHub repository savasturkoglu1/promptpilot-plugin (1 stars, last pushed 9d ago), licensed MIT. It adds 63 tokens to every session and 3,110 once invoked, about $0.0003 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
coding-agents-prompt-authoring
To author, adapt, review, and validate prompts (skills, agents, workflows, rules, etc.) with brief, contracts, and a validation pack.
prompt-tuning
Tune a prompt, or anything whose quality is measured by non-deterministic model output, without chasing noise - a noise baseline before the first edit, medians over repeated runs, enforcement AFTER generation rather than in the wording. Use when iterating on prompts or model-judged output.
composer-prompting
Internal guidance for shaping a well-specified coding task into a tight Cursor/Composer prompt before delegating it via /cursor:delegate.
ai-evaluation-engineering
An AI evaluation engineering specialist for testing models and prompts. The description does not provide enough detail to state which concrete operations it performs.
context-engineering
Optimize Claude Code context-window usage for accuracy and cost. TRIGGER when: hitting context limits, structuring prompts for an agent, or trimming what gets injected into a task. SKIP: persisting knowledge across sessions (use agent-memory); semantic recall tuning (use semantic-memory-mcp).
engineer-prompts
Builds or audits testable prompt contracts with explicit outcomes, permissions, tools, evidence, and stop conditions. Use when writing reusable agent prompts, system prompts, or prompts with unclear success criteria.