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/apurvbazari/claude-plugins/createnpx skills add ApurvBazari/claude-plugins --skill creategit clone --depth 1 https://github.com/ApurvBazari/claude-pluginsWhat 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 | $0.00111 | $0.02084 |
| Opus 5 | $0.00056 | $0.01042 |
| Sonnet 5 | $0.00022 | $0.00417 |
| Haiku 4.5 | $0.00011 | $0.00208 |
Grade A, and why
create 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create — Render the Session as an Interactive Document
You are invoked via /walkthrough:create [optional focus], or auto-invoked when the user asks
to visualize/recap the session. Produce ONE self-contained interactive HTML file in the house
style, then offer to open it. The renderer is six references/ files (including concept-coverage.md)
plus the references/components/ catalog (its index.md routes to per-group recipe files you read on
demand). Load them as the steps direct.
Step 1: Scope
If an argument was given (e.g. architecture decisions), scope synthesis to that focus.
Otherwise cover the whole session. If the session is trivially short, ask via AskUserQuestion
(2 fixed options: "Generate anyway" / "Skip") before proceeding.
Step 2: Gather
The session transcript is the source of record — synthesize from the conversation, not from
repository state. Read any source files you will cite so code locations (path:line) are real —
never invent a line number; if unverified, cite path only or omit.
Step 3: Synthesize the session model
Build the structured model per references/session-model.md (title, summary, typeTags,
sections[], nodes[], edges[], decisions[], files[], timeline[], metrics[], openQuestions[],
details{}) BEFORE writing any HTML.
Step 4: Coverage critic
Run references/completeness.md Part 1 against the session before selecting components. Fold omitted
salient items into the model; note intentional omissions for the coverage note.
Step 5: Select components
Using references/authoring-guide.md, map the model to component names, then look each name up in
references/components/index.md to find its group file. Apply "omit empty, never stub". For content
that fits no catalog entry, compose a bespoke component per the authoring-guide recipe + looks-native checklist.
Before accepting a catalog diagram, run the authoring-guide § 1 diagram-fidelity check: cyclic /
guarded nodes[]+edges[] → state / transition diagram; timed messages between actors → sequence /
swimlane diagram; a shape the catalog still cannot draw faithfully → compose a bespoke diagram. Never
force-fit a state machine or a message trace into a flow / architecture map.
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.
- references/authoring-guide.md 16 KB
- references/completeness.md 2.9 KB
- references/components/data.md 8.6 KB
- references/components/decisions.md 9.8 KB
- references/components/diagrams.md 28 KB
- references/components/files-timeline.md 8.7 KB
- references/components/index.md 4.1 KB
- references/components/interactive.md 5.8 KB
- references/components/layout-prose.md 8.1 KB
- references/components/metrics.md 4.8 KB
- references/components/reasoning.md 3.3 KB
- references/components/review.md 7.2 KB
- references/concept-coverage.md 5.6 KB
- references/design-system.md 4.7 KB
- references/interactivity.md 12 KB
- references/page-scaffold.md 18 KB
- references/seed.html 31 KB
- references/self-check.md 7.8 KB
- references/session-model.md 16 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.
- 2d ago First seen · 130 lines · 111 tokens per session scan A 8246fbbcdc04
create is a skill published in the GitHub repository ApurvBazari/claude-plugins (0 stars, last pushed 26d ago), licensed MIT. It adds 111 tokens to every session and 2,084 once invoked, about $0.0006 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
frame-macos-notification
拟真 macOS 通知 banner + app icon + 标题正文, 适合 video overlay / 产品发布预告.
media-processing
Ingest and process media files (video, audio, image).
visualize
Render a polished visual inline in the chat as part of your answer — a diagram, a chart, an interactive explainer, or a UI mockup. Load it proactively whenever an explanation would land better as a picture than as prose. Do not wait to be asked.
fish-audio
Generate expressive audio clips using Fish Audio S2 TTS with bracket emotion tags. Record voice memos, narration, audio messages, or any spoken content.
image-studio
Create images from a text description, or edit photos and graphics the user provides (remove backgrounds or watermarks, retouch, restyle, in-paint). Can produce multiple variants when the user wants options to choose from.
baocut
BaoCut-only operator for the installed bcut CLI and .bcut projects. Implicitly trigger only when the request explicitly names BaoCut or bcut, targets a .bcut project or BaoCut Subtitle Studio, or continues a BaoCut workflow already established in the conversation. Do not trigger solely for generic audio or video…