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/mblode/ghostwriter/evaluate-ghostwriternpx skills add mblode/ghostwriter --skill evaluate-ghostwritergit clone --depth 1 https://github.com/mblode/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 | $0.00098 | $0.01656 |
| Opus 5 | $0.00049 | $0.00828 |
| Sonnet 5 | $0.00020 | $0.00331 |
| Haiku 4.5 | $0.00010 | $0.00166 |
Grade A, and why
evaluate-ghostwriter 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 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.
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 — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Evaluate ghostwriter
Measure a fixed runtime skill and fixed private profiles against real held-out writing.
- IS: clean candidate generation, deterministic blinding, human review, and descriptive reporting.
- IS NOT: training profiles, modifying corpora, grading with another model, or exposing the treatment before a human choice. Use
train-ghostwriterto create profiles and held-out cases.
The scripts execute the deterministic parts of this workflow. Do not reproduce their logic manually. scripts/run-eval.ts generates candidates without opening references. scripts/review-eval.ts joins references only after generation and records human choices. The scripts use assets/candidate-output.schema.json for structured runner output; do not edit it per run.
Both branches of a pair get the same CLI, model, case bytes, and output contract in fresh non-persistent sessions. The treatment alone also receives the complete runtime SKILL.md and platform profile, encoded losslessly as JSON strings. Never summarize or selectively copy either file; manifest.json pins hashes of their original bytes.
This measures the whole ghostwriter skill (anti-AI-prose pass plus strategy layer) and profile bundle against a raw-model baseline with no style guidance. A treatment win cannot be attributed to the profile alone, since it does not isolate the profile's own marginal effect.
Workflow
Copy this checklist and work top to bottom; each item is a section below.
- 1. Pin inputs and runner (cases, runtime skill, profiles dir, runs dir, runner, model)
- 2. Preview the provider boundary and get confirmation
- 3. Generate matched pairs with
run-eval.ts - 4. Review blind with
review-eval.ts, choosinga/b/tie/invalidper case - 5. Verify counts reconcile against
labels.jsonland preserve the immutable run
1. Pin inputs and runner
Resolve these paths explicitly:
evals/cases.jsonl, containing no held-out responses- the exact installed
ghostwriter/SKILL.md - the private profile directory
soul.mdin the profiles dir (optional, cross-platform voice core), sent with the treatment when present and hashed into the manifest assoulHashevals/runs- one runner (
codexorclaude) and an explicit model
What ships with it
3 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.
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 · 111 lines · 98 tokens per session scan A 1c8b0cfb0c81
evaluate-ghostwriter is a skill published in the GitHub repository mblode/ghostwriter (7 stars, last pushed 27d ago), licensed MIT. It adds 98 tokens to every session and 1,656 once invoked, about $0.0005 per session on Opus 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
agent-host-chat-contributions
Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.