owenob1/prompt-improver / skills/prompt-improver
prompt-improver
VerifiedTransform vague prompts into precise, verifiable structured XML prompts that coding agents execute reliably. Modes: execute (default — generate then run) and plan (generate XML for review first). Use when the user says improve prompt, make this work better, prompt engineer, structure a request, plan a complex change before coding, or when a rough request needs verification criteria before execution. Do not use when the input is already a well-structured XML prompt or detailed implementation spec — skip generation and execute directly.
config/runtime-defaults.json
View on GitHub{
"version": 1,
"model_aliases": {
"mythos-5.1": "claude-mythos-5-1",
"mythos5.1": "claude-mythos-5-1",
"claude-mythos-5-1": "claude-mythos-5-1",
"mythos": "claude-mythos-5-1",
"mythos-5": "claude-mythos-5",
"mythos5": "claude-mythos-5",
"claude-mythos-5": "claude-mythos-5",
"mythos-preview": "claude-mythos-preview",
"claude-mythos-preview": "claude-mythos-preview",
"fable-5.1": "claude-fable-5-1",
"fable5.1": "claude-fable-5-1",
"claude-fable-5-1": "claude-fable-5-1",
"fable-5": "claude-fable-5",
"fable5": "claude-fable-5",
"claude-fable-5": "claude-fable-5",
"fable": "fable",
"opus-5.5": "claude-opus-5-5",
"opus5.5": "claude-opus-5-5",
"claude-opus-5-5": "claude-opus-5-5",
"claude-opus-5.5": "claude-opus-5-5",
"opus-5": "claude-opus-5",
"opus5": "claude-opus-5",
"claude-opus-5": "claude-opus-5",
"opus-4.8": "claude-opus-4-8",
"opus4.8": "claude-opus-4-8",
"claude-opus-4-8": "claude-opus-4-8",
"opus-4.7": "claude-opus-4-7",
"opus4.7": "claude-opus-4-7",
"claude-opus-4-7": "claude-opus-4-7",
"opus-4.6": "claude-opus-4-6",
"opus4.6": "claude-opus-4-6",
"claude-opus-4-6": "claude-opus-4-6",
"opus": "opus",
"sonnet-5": "claude-sonnet-5",
"sonnet5": "claude-sonnet-5",
"claude-sonnet-5": "claude-sonnet-5",
"sonnet-4.6": "claude-sonnet-4-6",
"sonnet4.6": "claude-sonnet-4-6",
"claude-sonnet-4-6": "claude-sonnet-4-6",
"sonnet": "sonnet",
"haiku-4.5": "claude-haiku-4-5",
"haiku4.5": "claude-haiku-4-5",
"claude-haiku-4-5": "claude-haiku-4-5",
"claude-haiku-4.5": "claude-haiku-4-5",
"claude-haiku-4-5-20251001": "claude-haiku-4-5-20251001",
"haiku": "haiku",
"gpt-6-sol": "gpt-6-sol",
"gpt6-sol": "gpt-6-sol",
"sol": "gpt-6-sol",
"gpt-6": "gpt-6-sol",
"gpt6": "gpt-6-sol",
"gpt-6-astra": "gpt-6-astra",
"gpt6-astra": "gpt-6-astra",
"astra": "gpt-6-astra",
"gpt-6-luna": "gpt-6-luna",
"gpt6-luna": "gpt-6-luna",
"luna": "gpt-6-luna",
"gpt-5.6-sol": "gpt-5.6-sol",
"gpt5.6-sol": "gpt-5.6-sol",
"gpt-5.6": "gpt-5.6-sol",
"gpt5.6": "gpt-5.6-sol",
"gpt-5.6-terra": "gpt-5.6-terra",
"gpt5.6-terra": "gpt-5.6-terra",
"terra": "gpt-5.6-terra",
"gpt-5.6-luna": "gpt-5.6-luna",
"gpt5.6-luna": "gpt-5.6-luna",
"gpt-5.5": "gpt-5.5",
"gpt5.5": "gpt-5.5",
"gpt-5": "gpt-5.5",
"gpt5": "gpt-5.5",
"gpt-5.3-codex": "gpt-5.3-codex",
"gpt5.3-codex": "gpt-5.3-codex",
"gpt-5.2-codex": "gpt-5.2-codex",
"gpt5.2-codex": "gpt-5.2-codex",
"codex": "gpt-6-sol",
"openai": "gpt-6-sol",
"o4-mini": "o4-mini",
"o4mini": "o4-mini",
"grok-4.7": "grok-4.7",
"grok4.7": "grok-4.7",
"grok": "grok-4.7",
"grok-4.6": "grok-4.6",
"grok4.6": "grok-4.6",
"grok-4.5": "grok-4.5",
"grok4.5": "grok-4.5",
"grok-4.3": "grok-4.3",
"grok4.3": "grok-4.3",
"grok-build-0.1": "grok-build-0.1",
"grok-build": "grok-build-0.1",
"grokbuild": "grok-build-0.1",
"grok-code-fast-1": "grok-build-0.1",
"grok-code-fast": "grok-build-0.1",
"composer-2.5-fast": "grok-build-0.1",
"composer2.5-fast": "grok-build-0.1",
"grok-composer-2.5-fast": "grok-build-0.1",
"composer-2.5": "grok-build-0.1",
"composer2.5": "grok-build-0.1",
"grok-composer-2.5": "grok-build-0.1",
"gemini-3.8-flash": "gemini-3.8-flash",
"gemini3.8-flash": "gemini-3.8-flash",
"gemini-3.7-flash": "gemini-3.7-flash",
"gemini3.7-flash": "gemini-3.7-flash",
"gemini-3.6-flash": "gemini-3.6-flash",
"gemini3.6-flash": "gemini-3.6-flash",
"gemini-3.5-flash": "gemini-3.5-flash",
"gemini3.5-flash": "gemini-3.5-flash",
"gemini-3.5-flash-lite": "gemini-3.5-flash-lite",
"gemini3.5-flash-lite": "gemini-3.5-flash-lite",
"gemini-3.1-flash-lite": "gemini-3.1-flash-lite",
"gemini3.1-flash-lite": "gemini-3.1-flash-lite",
"gemini-3.1-pro-preview": "gemini-3.1-pro-preview",
"gemini-3.1-pro": "gemini-3.1-pro-preview",
"gemini3.1-pro": "gemini-3.1-pro-preview",
"gemini-pro": "gemini-3.1-pro-preview",
"gemini-2.5-pro": "gemini-2.5-pro",
"gemini2.5-pro": "gemini-2.5-pro",
"gemini-2.5-flash": "gemini-2.5-flash",
"gemini2.5-flash": "gemini-2.5-flash",
"gemini-flash": "gemini-3.8-flash",
"gemini": "gemini-3.8-flash"
},
"model_backend_patterns": [
{
"patterns": [
"mythos*",
"fable*",
"claude-*",
"sonnet*",
"haiku*",
"opus*",
"claude"
],
"backend": "claude"
},
{
"patterns": [
"grok*",
"composer*"
],
"backend": "grok"
},
{
"patterns": [
"gemini*"
],
"backend": "gemini"
},
{
"patterns": [
"gpt*",
"o1*",
"o3*",
"o4*",
"codex*",
"openai",
"chatgpt*",
"sol",
"terra",
"luna",
"astra"
],
"backend": "codex"
}
],
"model_fallback_chains": [
{
"//": "Every chain starts with the requested id; the rest are same-CLI fallbacks.",
"patterns": [
"*mythos*"
],
"chain": [
"$primary",
"claude-mythos-5-1",
"claude-mythos-5",
"fable",
"opus",
"sonnet"
]
},
{
"patterns": [
"*fable*"
],
"chain": [
"$primary",
"fable",
"opus",
"sonnet"
]
},
{
"patterns": [
"*opus*"
],
"chain": [
"$primary",
"opus",
"sonnet"
]
},
{
"patterns": [
"*sonnet*"
],
"chain": [
"$primary",
"sonnet"
]
},
{
"patterns": [
"*haiku*"
],
"chain": [
"$primary",
"haiku",
"sonnet"
]
},
{
"patterns": [
"gpt-6*"
],
"chain": [
"$primary",
"gpt-6-sol",
"gpt-6-luna",
"gpt-5.6-terra",
"gpt-5.5"
]
},
{
"patterns": [
"gpt-5.6*"
],
"chain": [
"$primary",
"gpt-6-sol",
"gpt-5.6-terra",
"gpt-5.6-luna",
"gpt-5.5"
]
},
{
"//": "gpt-5.5 retires from Codex on 2026-10-14; older codex ids are deprecated",
"patterns": [
"*codex*",
"gpt-5*",
"o[0-9]*"
],
"chain": [
"$primary",
"gpt-6-sol",
"gpt-6-luna"
]
},
{
"patterns": [
"grok-4*"
],
"chain": [
"$primary",
"grok-4.7",
"grok-4.6",
"grok-4.5"
]
},
{
"patterns": [
"grok*"
],
"chain": [
"$primary",
"grok-4.7"
]
},
{
"patterns": [
"*gemini*pro*"
],
"chain": [
"$primary",
"gemini-3.1-pro-preview",
"gemini-3.8-flash",
"gemini-2.5-pro"
]
},
{
"patterns": [
"*gemini*flash*"
],
"chain": [
"$primary",
"gemini-3.8-flash",
"gemini-3.7-flash",
"gemini-2.5-flash"
]
},
{
"patterns": [
"gemini*"
],
"chain": [
"$primary",
"gemini-3.8-flash"
]
}
],
"supported_backends": [
"claude",
"grok",
"gemini",
"codex",
"agy",
"copilot",
"cursor",
"opencode",
"cline",
"qwen",
"droid",
"amp",
"kimi",
"kiro"
],
"backend_commands": {
"//": "Used when backend_invocation=commands (or auto with your own override). {prompt_file} {model} {model_args} {max_tokens} are substituted; {model} is shell-quoted. Scripts in scripts/backends/ add timeouts and large-prompt handling that these templates cannot.",
"claude": "claude -p \"$(cat \"{prompt_file}\")\" --tools \"\" --output-format text --no-session-persistence --permission-mode dontAsk {model_args} </dev/null",
"grok": "grok -p \"$(cat \"{prompt_file}\")\" --output-format plain --always-approve --no-subagents --no-plan --disable-web-search --max-turns 3 {model_args} </dev/null",
"gemini": "gemini -p \"$(cat \"{prompt_file}\")\" --output-format text --approval-mode default {model_args} </dev/null",
"codex": "codex exec --sandbox read-only --skip-git-repo-check --ephemeral --color never {model_args} - <\"{prompt_file}\"",
"agy": "agy -p \"$(cat \"{prompt_file}\")\" --output-format text {model_args} </dev/null",
"copilot": "copilot -p \"$(cat \"{prompt_file}\")\" -s --no-ask-user --deny-tool shell --deny-tool write {model_args} </dev/null",
"cursor": "cursor-agent -p \"$(cat \"{prompt_file}\")\" --mode ask --output-format text {model_args} </dev/null",
"opencode": "opencode run {model_args} \"$(cat \"{prompt_file}\")\" </dev/null",
"cline": "cline --plan -y {model_args} \"$(cat \"{prompt_file}\")\" </dev/null",
"qwen": "qwen -p \"$(cat \"{prompt_file}\")\" --output-format text --approval-mode default {model_args} </dev/null",
"droid": "droid exec -f \"{prompt_file}\" {model_args} </dev/null",
"amp": "amp -x <\"{prompt_file}\"",
"kimi": "kimi -p \"$(cat \"{prompt_file}\")\" --output-format text {model_args} </dev/null",
"kiro": "kiro-cli chat --no-interactive {model_args} \"$(cat \"{prompt_file}\")\" </dev/null"
},
"backend_model_flags": {
"claude": "--model {model}",
"grok": "-m {model}",
"gemini": "-m {model}",
"codex": "-m {model}",
"agy": "--model {model}",
"copilot": "--model {model}",
"cursor": "--model {model}",
"opencode": "-m {model}",
"cline": "-m {model}",
"qwen": "-m {model}",
"droid": "-m {model}",
"amp": "",
"kimi": "-m {model}",
"kiro": "--model {model}"
},
"limit_detection": {
"account_patterns": "weekly.?limit|monthly.?limit|session limit|spend limit|hit your .*limit|you.?ve hit your|you have hit your|you.?ve reached your|organization.?limit|org.?limit|account.?limit|individual usage limit|out of (usage|credits)|usage.?limit.?reached|credit balance is too low|exhausted your daily quota|grok build usage limit|upgrade to (plus|pro|supergrok)|limit · resets|limit · reset",
"retry_patterns": "rate.?limit|usage.?limit|quota|out of (limit|usage|credits)|capacity|overloaded|(^|[^0-9])(401|403|429|503|529)([^0-9]|$)|not (available|supported|enabled) (for|on|with|in)|model .*not found|unknown model|invalid model|not a (recognized|valid) model|model .* (denied|restricted|not accessible)|restricted by your organization|access denied|does not have access|not available with the|invitation|glasswing|temporarily (unavailable|limiting)|service unavailable|try again (later|at)|too many requests|resource.?exhausted|throttl|not supported when using codex",
"xml_markers": "^[[:space:]]*<[A-Za-z][A-Za-z0-9_-]*[[:space:]>/]"
},
"host_env_markers": {
"claude": [
"CLAUDECODE",
"CLAUDE_CODE_ENTRYPOINT",
"CLAUDE_CODE",
"CLAUDE_AGENT"
],
"grok": [
"GROK_BUILD",
"XAI_GROK",
"GROK_SESSION"
],
"gemini": [
"GEMINI_CLI",
"GOOGLE_GEMINI_CLI"
],
"codex": [
"CODEX_SANDBOX",
"CODEX_MANAGED_BY_NPM",
"OPENAI_CODEX"
],
"qwen": [
"QWEN_CODE"
],
"//": "Only variables a CLI is known to export to its child shells. Checked in supported_backends order. CODEX_HOME is deliberately absent: it is user config, not a session marker."
},
"parent_process_patterns": {
"claude": [
"claude",
"claude-*",
"*claude*"
],
"grok": [
"grok",
"grok-*",
"*grok*"
],
"gemini": [
"gemini",
"gemini-*",
"*gemini*"
],
"codex": [
"codex",
"codex-*",
"*codex*"
],
"agy": [
"agy",
"antigravity*"
],
"copilot": [
"copilot",
"copilot-*"
],
"cursor": [
"cursor-agent",
"cursor-agent*"
],
"opencode": [
"opencode",
"opencode-*"
],
"cline": [
"cline",
"cline-*"
],
"qwen": [
"qwen",
"qwen-*"
],
"droid": [
"droid",
"droid-*"
],
"amp": [
"amp"
],
"kimi": [
"kimi",
"kimi-*"
],
"kiro": [
"kiro",
"kiro-*",
"kiro-cli*"
]
},
"cascade_scan_order": [
"claude",
"codex",
"grok",
"gemini",
"agy",
"copilot",
"cursor",
"opencode",
"cline",
"qwen",
"droid",
"amp",
"kimi",
"kiro"
],
"generation": {
"context_mode": "deterministic",
"include_xml_template": true,
"include_principles": true,
"include_chaining": true,
"include_examples": true,
"include_system_prompt": true,
"system_prompt_path": "assets/generation-agent-prompt.md",
"xml_template_path": "references/xml-template.md",
"principles_path": "references/prompting-principles.md",
"chaining_path": "references/prompt-chaining.md",
"examples_path": "examples/before-after.md",
"extra_reference_paths": [],
"output_instructions": "Output ONLY the final improved XML prompt. No explanation, no code fences, no commentary, no execution of the request.",
"require_xml_output": true,
"forbid_agent_codebase_search": true,
"backend_timeout_secs": 300,
"grok_timeout_secs": 180,
"grok_max_turns": 3
},
"backend_binaries": {
"//": "CLI executable(s) per backend, first found wins. Default: the backend name.",
"cursor": [
"cursor-agent",
"agent"
],
"kiro": [
"kiro-cli",
"kiro"
]
}
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |