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 agentmods add skills/an2n/ghostwriter/checknpx skills add an2n/ghostwriter --skill checkgit clone --depth 1 https://github.com/an2n/ghostwriterWhat 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 | $0.00079 | $0.05229 |
| Opus 5 | $0.00039 | $0.02615 |
| Sonnet 5 | $0.00016 | $0.01046 |
| Haiku 4.5 | $0.00008 | $0.00523 |
Grade A, and why
check scanned grade A with 1 finding 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 2d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- Curly quotes alone - many editors and CMSes auto-curl. How it starts
The opening of the file, as written. The whole thing — 409 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Check
Reads a piece of text and reports how strongly it reads as AI-written, with a concrete
list of what fired and why. This is a find-only skill - it never rewrites anything. See
polish for the fix side.
How the scoring works
Every check below is a tell: something that either fires or doesn't. Each tell carries a fixed weight - 1 (weak), 2 (moderate), or 3 (strong) - based on how reliable it is as evidence on its own. Add up the weights of everything that fires, then divide by the text's length in hundreds of words to get a tell density: points per 100 words. A 2,000-word article and a 40-word reply shouldn't be judged on the same raw total, so density normalizes for length. For anything under 100 words - the common case for a single post or reply - skip the division and use the raw total directly against the same bands, with confidence capped lower given the small sample.
| Density (points per 100 words) | Verdict |
|---|---|
| 0 - 1 | Reads human |
| 1 - 3 | Mostly human |
| 3 - 6 | Mixed signals |
| 6 - 10 | Reads AI |
| 10+ | Heavily AI |
A phrase can count toward at most two tells when it's genuinely diagnostic for both (e.g. "it is important to note that" is both stock vocabulary and reflexive hedging) - never double-count it twice within the same tell.
Cluster 1: Word choice
Vocabulary and phrase-level tells - wrong regardless of what sentence they sit in.
Stock AI vocabulary (weight 1 each, capped at 3 points total for this tell even if more fire): delve, leverage (verb), utilize, robust, comprehensive, streamline, foster/fostering, facilitate, pivotal, nuanced, notable/notably, enduring, garner, valuable, vibrant, intricate/intricacies, interplay, tapestry (abstract noun), testament (figurative), underscore (verb), showcase (verb), key (as a vague adjective), align with, emphasizing, enhance, quietly (as a hedge-flavored adverb), multifaceted, in the realm of, the landscape of (abstract noun), a myriad of, a plethora of, actually (as filler), additionally (as an opener), it is worth noting, it is important to note. Cite the exact word that fired.
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.
- 2d ago First seen · 409 lines · 79 tokens per session scan A 1c9ed33233f3
check is a skill published in the GitHub repository an2n/ghostwriter (1 stars, last pushed 6d ago), licensed MIT. It adds 79 tokens to every session and 5,229 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
writing-antipatterns
When editing, creating, writing content and articles to avoid AI writing patterns ("AI-isms") that make text sound machine-generated. Follow the guidelines in the skill description to audit, rewrite, and summarize changes.
apm-usage
Activate when the user asks about APM (Agent Package Manager): installing, configuring, authoring, or troubleshooting AI-agent packages, dependencies, compilation, MCP servers, policy, or any apm CLI command.
clean-user-facing-text
Audit and finalize authorized natural-language text meant for readers: strip suspicious invisible Unicode, then rewrite prose while keeping facts, meaning, and the writer's voice. Use when the user asks to clean, humanize, polish, or finalize articles, manuscripts, reports, documentation, emails, product copy, UI…
patina
Detect and rewrite AI writing patterns in Korean, English, Chinese, and Japanese text so it reads as if a human wrote it. Meaning-preservation (MPS) verified.
sessions
Search and ask questions about coding agent session history across Claude Code, Codex, and Cursor. Use when asking what was worked on, what was tried before, how a problem was investigated across sessions, what happened recently, or any question about past agent sessions. Also use when the user references prior…
codex-autoresearch
Run autonomous, measurable experiments in a Git repository: change one hypothesis, verify a numeric metric, keep improvements, and revert failures. Use when the user wants Codex to keep iterating toward a numeric target in the foreground or as a detached background run. Do not use for ordinary one-shot coding…