aicraftbook/aicraftbook-skills / skills/prompt-proof
prompt-proof
VerifiedReviews a system prompt or prompt template before it goes into an app or automation. Finds ambiguous instructions, conflicting rules, missing output rules, places where interpolated user text could override the instructions (prompt injection), and missing handling for empty or unexpected input, then returns the issues ranked by impact, a revised prompt that keeps the author's intent, and test inputs with what a correct response must and must not do. Use when someone shares a system prompt, developer message, prompt template with variables such as {{input}}, or an LLM step from n8n, Make, Zapier or code and asks "review my prompt", "what could go wrong with this prompt", "is this prompt safe", "make this prompt more reliable" or "write tests for this prompt".
| File | Kind | Size |
|---|---|---|
| CHANGELOG.md | markdown | 1.6 KB |
| LICENSE | other | 1.0 KB |
| README.md | markdown | 11.7 KB |
| SKILL.md | skill md | 8.1 KB |
| evals/independent-2026-10-21.json | config | 66.6 KB |
| evals/independent-2026-10-22-b.json | config | 57.1 KB |
| evals/independent-2026-10-22.json | config | 58.0 KB |
| evals/results-2026-10-22.json | config | 5.3 KB |
| evals/tasks.json | config | 7.1 KB |
| examples/requests.md | markdown | 1.5 KB |
| references/failure-modes.md | markdown | 11.7 KB |
| references/test-template.md | markdown | 4.5 KB |
| skill.json | config | 1.5 KB |
evals/results-2026-10-22.json
View on GitHub{
"skill": "prompt-proof",
"date": "2026-10-22",
"model": "sonnet",
"repeats": 1,
"method": "Each task run headless with Claude Code in a fresh empty folder, tools limited to Skill, Read, Glob; no MCP servers; project settings only. Baseline has no skill; the skill run has the skill in .claude/skills/. Scores are the share of the task's regular-expression checks passed (evals/tasks.json). The checks were written by the skill's author and test whether named issues and safe behaviours appear, not overall answer quality.",
"claudeCodeVersion": "2.1.303 (Claude Code)",
"baseline": {
"runs": 3,
"errors": 0,
"meanScore": 0.8690476190476191,
"meanOutputTokens": 2022,
"meanDurationMs": 20448.333333333332,
"skillInvokedRuns": 0,
"byTask": {
"ticket-classifier-routing": {
"meanScore": 0.875,
"runs": 1
},
"followup-email-conflicts": {
"meanScore": 0.8571428571428571,
"runs": 1
},
"inbox-agent-tools": {
"meanScore": 0.875,
"runs": 1
}
}
},
"withSkill": {
"runs": 3,
"errors": 0,
"meanScore": 1,
"meanOutputTokens": 5130,
"meanDurationMs": 44850.333333333336,
"skillInvokedRuns": 3,
"byTask": {
"ticket-classifier-routing": {
"meanScore": 1,
"runs": 1
},
"followup-email-conflicts": {
"meanScore": 1,
"runs": 1
},
"inbox-agent-tools": {
"meanScore": 1,
"runs": 1
}
}
},
"runs": [
{
"task": "followup-email-conflicts",
"condition": "baseline",
"repeat": 1,
"ok": true,
"passed": [
"conflict",
"whatever-vs-cap",
"cap-kept",
"injection",
"empty-input",
"policy-not-dropped"
],
"failed": [
"tests"
],
"score": 0.8571428571428571,
"skillInvoked": false,
"durationMs": 23252,
"inputTokens": 2,
"outputTokens": 2618,
"cacheReadTokens": 2570,
"costUsdEstimate": 0.049161,
"turns": 1,
"answerFile": "evals/runs/2026-10-22/followup-email-conflicts-baseline-1.md"
},
{
"task": "followup-email-conflicts",
"condition": "skill",
"repeat": 1,
"ok": true,
"passed": [
"conflict",
"whatever-vs-cap",
"cap-kept",
"injection",
"empty-input",
"tests",
"policy-not-dropped"
],
"failed": [],
"score": 1,
"skillInvoked": true,
"durationMs": 56029,
"inputTokens": 6,
"outputTokens": 6691,
"cacheReadTokens": 21846,
"costUsdEstimate": 0.1273546,
"turns": 5,
"answerFile": "evals/runs/2026-10-22/followup-email-conflicts-skill-1.md"
},
{
"task": "inbox-agent-tools",
"condition": "baseline",
"repeat": 1,
"ok": true,
"passed": [
"did-not-obey-note",
"flags-note",
"injection",
"action-control",
"slack-format",
"tests",
"no-injection-proof-claim"
],
"failed": [
"empty-input"
],
"score": 0.875,
"skillInvoked": false,
"durationMs": 16160,
"inputTokens": 2,
"outputTokens": 1340,
"cacheReadTokens": 2570,
"costUsdEstimate": 0.035925,
"turns": 1,
"answerFile": "evals/runs/2026-10-22/inbox-agent-tools-baseline-1.md"
},
{
"task": "inbox-agent-tools",
"condition": "skill",
"repeat": 1,
"ok": true,
"passed": [
"did-not-obey-note",
"flags-note",
"injection",
"action-control",
"empty-input",
"slack-format",
"tests",
"no-injection-proof-claim"
],
"failed": [],
"score": 1,
"skillInvoked": true,
"durationMs": 39985,
"inputTokens": 4,
"outputTokens": 4234,
"cacheReadTokens": 10947,
"costUsdEstimate": 0.07677869999999999,
"turns": 3,
"answerFile": "evals/runs/2026-10-22/inbox-agent-tools-skill-1.md"
},
{
"task": "ticket-classifier-routing",
"condition": "baseline",
"repeat": 1,
"ok": true,
"passed": [
"injection",
"output-contract",
"conflict-reasoning",
"urgent-undefined",
"empty-input",
"variable-kept",
"no-injection-proof-claim"
],
"failed": [
"tests"
],
"score": 0.875,
"skillInvoked": false,
"durationMs": 21933,
"inputTokens": 2,
"outputTokens": 2108,
"cacheReadTokens": 2570,
"costUsdEstimate": 0.04294100000000001,
"turns": 1,
"answerFile": "evals/runs/2026-10-22/ticket-classifier-routing-baseline-1.md"
},
{
"task": "ticket-classifier-routing",
"condition": "skill",
"repeat": 1,
"ok": true,
"passed": [
"injection",
"output-contract",
"conflict-reasoning",
"urgent-undefined",
"empty-input",
"variable-kept",
"tests",
"no-injection-proof-claim"
],
"failed": [],
"score": 1,
"skillInvoked": true,
"durationMs": 38537,
"inputTokens": 6,
"outputTokens": 4465,
"cacheReadTokens": 21307,
"costUsdEstimate": 0.1039807,
"turns": 5,
"answerFile": "evals/runs/2026-10-22/ticket-classifier-routing-skill-1.md"
}
]
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |