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 robertguss/claude-code-toolkit --skill ghost-writergit clone --depth 1 https://github.com/robertguss/claude-code-toolkitWrote 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/robertguss/claude-code-toolkit/ghost-writer)<a href="https://agentmods.dev/skills/robertguss/claude-code-toolkit/ghost-writer"><img src="https://agentmods.dev/badge/skills/robertguss/claude-code-toolkit/ghost-writer/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/robertguss/claude-code-toolkit/ghost-writer"><img src="https://agentmods.dev/badge/skills/robertguss/claude-code-toolkit/ghost-writer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Prompt Injection · line 91 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 98 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 99 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
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 | $0.00077 | $0.03753 |
| Opus 5.5 | $0.00031 | $0.01501 |
| Sonnet 5.5 | $0.00015 | $0.00751 |
| Haiku 4.5 | $0.00008 | $0.00375 |
Grade A, and why
ghost-writer 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 — 497 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ghost Writer
Produce first drafts at ~80% voice accuracy using a writer's Voice DNA Document.
Core Philosophy
You are a collaborative writing partner, not an order-taker.
- Evaluate, don't just accept — Assess task clarity, research sufficiency, and DNA-task fit. If something seems off, say so.
- Surface tensions proactively — DNA vs. task conflicts, potential issues, gaps in research or direction.
- Offer honest feedback — On drafts, on approach, on choices made. The user benefits from your perspective.
- Push back diplomatically — When you see problems, raise them with reasoning. "I can do this, but here's a concern..."
- Advocate for quality — Note concerns while respecting user autonomy. If they insist after pushback, proceed faithfully.
- Share perspective even when not asked — You're a partner, not a tool. Offer observations proactively.
The user always decides. After pushback, if they say "proceed anyway," you do—noting the concern, then executing faithfully.
What This Skill Does
- Consumes Voice DNA Documents (full document, not just briefing section)
- Generates 2 meaningfully different first drafts
- Provides 2-3 headline options per draft
- Assesses confidence based on profile readiness and freshness
- Documents decisions made and reasoning
- Collects structured feedback and suggests DNA refinements
- Supports iteration until the user is satisfied
Dependencies
Voice DNA Document Required
This skill requires a Voice DNA Document as input every session. The document should be produced by the writing-dna-discovery skill, containing:
- Voice Profile (sentence patterns, punctuation, word choice, tone, reader relationship)
- Ghost Writer Briefing (Do This, Don't Do This, When Uncertain)
- Exemplar Passages (annotated examples)
- Anti-Patterns (what to avoid)
- Readiness Level (Minimum Viable, Solid, or Strong)
If no DNA document is provided, do not proceed. Direct the user to the writing-dna-discovery skill first.
What ships with it
19 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- design-notes.md 15 KB
- references/anti-ai-patterns.md 11 KB
- references/blog-writing-guide.md 7.2 KB
- references/clarity-and-cognition.md 8.9 KB
- references/closing-strategies.md 9.1 KB
- references/common-writing-weaknesses.md 11 KB
- references/elements-of-style.md 11 KB
- references/feedback-collection-protocol.md 9.2 KB
- references/long-form-essay-guide.md 8.5 KB
- references/on-writing-well.md 8.4 KB
- references/opening-strategies.md 10 KB
- references/output-format-guide.md 12 KB
- references/platform-conventions.md 8.1 KB
- references/quality-checklist.md 8.2 KB
- references/sentence-mastery.md 9.4 KB
- references/session-flow-guide.md 12 KB
- references/transition-mastery.md 7.8 KB
- references/voice-calibration-techniques.md 8.4 KB
- references/voice-consumption-guide.md 12 KB
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 First seen · 497 lines · 77 tokens per session scan A 214be3066514
ghost-writer is a skill published in the GitHub repository robertguss/claude-code-toolkit (126 stars, last pushed 1mo ago), licensed MIT. It adds 77 tokens to every session and 3,753 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
offensive-vuln-classes
Exploit development curriculum covering core vulnerability classes with real-world CVE case studies: stack/heap buffer overflows, use-after-free, integer overflows, format strings, type confusion, and race conditions. Use when learning or teaching vuln classes, researching specific CVE patterns, or building exploit…
offensive-edr-evasion
EDR evasion offensive checklist: hook unhooking (user/kernel), direct syscalls, PPID spoofing, process injection variants, AMSI bypass, ETW patching, memory encryption, and behavior-based evasion. Use when planning EDR bypass during red team engagements or researching AV/EDR evasion techniques.
offensive-mitigations
Security mitigation reference and bypass catalog: ASLR, DEP/NX, RELRO, stack canaries, CFI, sandboxing, seccomp. Covers both detection of enabled mitigations and known bypass techniques. Use when assessing target hardening or planning exploit mitigation bypasses.
offensive-initial-access
Initial access techniques checklist: phishing (spear/smishing), credential stuffing, exposed service exploitation, supply chain attacks, watering hole, VPN/RDP brute force, public-facing application exploitation. Maps to MITRE ATT&CK TA0001. Use when planning initial access phases of red team engagements.
offensive-exploit-development
Exploit development operational guide: environment setup, debugging workflow, PoC development lifecycle, writing reliable exploits, using pwntools/pwndbg, heap exploitation techniques, and weaponization considerations. Use when actively developing exploits or setting up an exploit dev environment.
offensive-container-escape
Container escape and breakout techniques targeting Docker, containerd, and Podman runtimes. Covers privileged container breakout via host filesystem mount and nsenter, Docker socket abuse through /var/run/docker.sock, Linux capability exploitation including CAPSYSADMIN, CAPSYSPTRACE, and CAPNETADMIN, cgroup v1…