Borrowing it
Nothing to install: this file belongs to lsampaioweb/ai-instructions. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/lsampaioweb/ai-instructions/main/.cursor/skills/review-ai-customization-files/SKILL.mdgit clone --depth 1 https://github.com/lsampaioweb/ai-instructionsWrote 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/lsampaioweb/ai-instructions/review-ai-customization-files)<a href="https://agentmods.dev/skills/lsampaioweb/ai-instructions/review-ai-customization-files"><img src="https://agentmods.dev/badge/skills/lsampaioweb/ai-instructions/review-ai-customization-files/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/lsampaioweb/ai-instructions/review-ai-customization-files"><img src="https://agentmods.dev/badge/skills/lsampaioweb/ai-instructions/review-ai-customization-files.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 | $0.00090 | $0.01978 |
| Opus 5.5 | $0.00036 | $0.00791 |
| Sonnet 5.5 | $0.00018 | $0.00396 |
| Haiku 4.5 | $0.00009 | $0.00198 |
Grade A, and why
review-ai-customization-files 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 — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Customization Audit
Arguments
- Required: file, file list, folder, or glob to audit.
- Optional for single-file scopes: explicit review-ledger path.
1. Scope and Analysis
- Review only the user-provided scope. If it is missing, stop and request it.
- Resolve the style contract in this order: workspace
.github/instructions/ai-customization.instructions.md, then the applicable user-level contract supplied by the environment. If neither exists, stop and report the missing prerequisite. - Load and apply the resolved style contract.
- Resolve the supplied scope to files. Require at least one readable target file; otherwise report the scope as invalid and return
BLOCKED. - Identify active references among the files in scope.
2. Review State and Convergence
- Load the review ledger before scanning when it exists.
- For a single-file scope, use
<file>.review-ledger.mdas the default ledger path. An explicit ledger path takes precedence over the default path. Validate and use the selected path consistently for loading, reporting, creation approval, and updates. If it does not exist, report it as not initialized and do not silently fall back to another path. If an existing ledger is unreadable or does not contain the required fields, reportReview ledger: invalidwith the path and reason, returnBLOCKED, and do not scan or modify the target file. - For a folder or glob, review each resolved file independently and use
<file>.review-ledger.mdbeside each file as its deterministic ledger path. Do not require one shared folder ledger or an explicit ledger path. - If the resolved ledger does not exist, report
Review ledger: not initializedand continue the current scan. Do not create an empty ledger. - Assign each new finding the next unused severity number in the applicable ledger or current report. Treat every new finding as
PROVISIONALuntil its ID and required fields are written to the ledger; only then mark its ID stateLEDGER. Preserve provisional IDs when the ledger is approved and created unless they conflict with existing entries. - Assign each finding a stable fingerprint based on canonical file identity, category, and normalized substantive problem. Do not include line location in the fingerprint.
- When a ledger exists, assign each new finding the next unused number for its severity in that ledger. Reuse the ledger ID when the same fingerprint appears again. Do not call a new finding immutable or
LEDGERuntil the ledger update is complete. Sort findings for display independently of their IDs. - Without a ledger, do not infer
FIXED,ACCEPTED, orREJECTEDstatus from conversation history, diffs, or filenames. Treat the review as having no persisted finding history. - Mark findings as
OPEN,FIXED,ACCEPTED, orREJECTED. - Do not report
FIXED,ACCEPTED, orREJECTEDfindings again unless the relevant content snapshot, referenced instruction, or applicable scope has materially changed. - When a resolved finding materially changes, retain its fingerprint and ID, mark it
OPEN, store the new snapshot, and append the transition to its resolution history. - A wording preference, optional rewrite, or alternative style is not a finding.
- When no new or unresolved substantive findings remain, report
Findings: noneand returnREADY. - When no ledger exists, state that current findings are clean only for this scan; do not claim that previous findings were resolved. If the scan has no findings, leave the ledger uncreated.
- Update the target files only after explicit user confirmation.
- When the user approves target-file fixes, create or update each affected file's adjacent ledger in the same operation and record the approved findings as
FIXED, unless the user explicitly excludes ledger changes. - Create or update the ledger only after explicit user confirmation when a finding is explicitly accepted or rejected without changing the target file. Do not create a ledger for a clean scan.
- Store at least the finding ID, fingerprint, file path, category, status, content snapshot or revision, and resolution note in the ledger. A valid ledger must be readable and contain those fields for every finding.
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 Changed · +35 lines · +31 tokens per session b983b2280880
- 2mo ago First seen · 86 lines · 59 tokens per session scan A 8a74432ebeda
review-ai-customization-files is a skill published in the GitHub repository lsampaioweb/ai-instructions (1 stars, last pushed 5d ago), licensed MIT. It adds 90 tokens to every session and 1,978 once invoked, about $0.0004 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 2026-08-31.
review-ai-customization-files compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| review-ai-customization-fileslsampaioweb/ai-instructions | Skill | 1 | 5d ago | MIT | A | 90 | Cursor |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 103 | Claude Code |