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/binary16labs/prime-silo/documentationnpx skills add binary16labs/prime-silo --skill documentationgit clone --depth 1 https://github.com/binary16labs/prime-siloWhat 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 | $0.00008 | $0.01386 |
| Opus 5 | $0.00004 | $0.00693 |
| Sonnet 5 | $0.00002 | $0.00277 |
| Haiku 4.5 | $0.00001 | $0.00139 |
Grade A, and why
Documentation 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.
This is a copy
92% identical to Documentation — 13 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Use this skill when you need project orientation before editing or answering architecture questions.
helper
- Import
/mod/_core/documentation/documentation.js read("path/to/file.md")reads a nested markdown doc relative todocs/url("path/to/file.md")builds the resolved/mod/...URL for a doc file
workflow
- Start with the built-in documentation index below unless you already know the exact doc path
- Use one focused
read("path/to/file.md")call instead of loading many large docs blindly - Treat
/README.mdas the public product source of truth for the project pitch, quick starts, release links, community links, and DeepWiki discovery - After orientation, confirm the concrete contract in the owning
AGENTS.mdfile and then inspect code when needed - For visual or modal work, confirm
app/L0/_all/mod/_core/visual/AGENTS.mdbefore changing dialog shells, buttons, cards, popovers, or other shared UI primitives - For dashboard panel creation, panel manifests, or panel-navigation helpers, start with
app/modules-and-extensions.md - Keep the repo's frontend-first rule in mind while reading: backend docs explain constraints and existing contracts, not default permission to edit
server/ - If the change appears to require backend work and the user did not explicitly ask for backend edits, ask for permission and explain the security, integrity, or stability reason before changing backend files
- When you change a stable contract or workflow, update both the relevant
AGENTS.mdfiles and the matching docs in/mod/_core/documentation/docs/ - When you add, remove, rename, or substantially repurpose a doc file, update this skill's in-file index in the same session
recommended starting points
- overall system shape:
architecture/overview.md - desktop host or packaging flow:
architecture/desktop-host-and-packaging.md - documentation rules:
architecture/documentation-system.md - frontend runtime and layers:
app/runtime-and-layers.md - admin agent runtime:
app/admin-agent-runtime.md - modules, routing, extensions, or dashboard panels:
app/modules-and-extensions.md - browser-side Hugging Face testing:
app/huggingface-browser-runtime.md - browser-side WebLLM testing:
app/webllm-browser-runtime.md - spaces and widgets:
app/spaces-and-widgets.md - overlay agent runtime:
agent/onscreen-agent-runtime.md - memory or promptinclude-backed agent memory:
agent/memory-and-prompt-includes.md - prompt or execution protocol:
agent/prompt-and-execution.md - skills or this documentation surface:
agent/skills-and-documentation.md - server routing or pages:
server/request-flow-and-pages.md - server jobs or maintenance loops:
server/jobs-and-maintenance.md - app-file APIs:
server/api/files.md - module, login, or runtime endpoints:
server/api/modules-and-runtime.md - auth and sessions:
server/auth-and-sessions.md - layered filesystem,
CUSTOMWARE_PATH, or writable-layer history:server/customware-layers-and-paths.md - CLI commands or runtime params:
cli/commands-and-runtime-params.md
docs path|name|description↓
architecture/overview.md|Runtime Overview|Browser-first architecture, major entry surfaces, and the layered runtime model.
architecture/desktop-host-and-packaging.md|Desktop Host And Packaging|Electron host startup, free-port binding, packaged single-user behavior, and desktop build outputs.
architecture/documentation-system.md|Documentation System|How AGENTS.md, the documentation module, and code fit together, plus update rules.
app/runtime-and-layers.md|App Runtime And Layers|Frontend boot flow, space runtime namespaces, entry shells, and L0/L1/L2 rules.
app/admin-agent-runtime.md|Admin Agent Runtime|Admin chat ownership, config persistence, shared execution loop, and API-versus-local-Hugging-Face transport switching.
app/modules-and-extensions.md|Modules And Extensions|/mod/... delivery, router path resolution, dashboard panel manifests, ext/html, ext/js, and <x-component> behavior.
app/huggingface-browser-runtime.md|Hugging Face Browser Runtime|The routed Transformers.js test surface, its worker split, direct Hub model loading contract, and throughput metrics.
app/webllm-browser-runtime.md|WebLLM Browser Runtime|The routed WebLLM test surface, its worker split, model-loading modes, and throughput metrics contract.
app/spaces-and-widgets.md|Spaces And Widgets|Space storage, widget renderer contracts, widget-shell defaults, and the main space.current / space.spaces helpers.
agent/onscreen-agent-runtime.md|Onscreen Agent Runtime|Overlay ownership, persistence, defaults, UI/runtime surfaces, and prompt file ownership.
agent/memory-and-prompt-includes.md|Memory And Prompt Includes|Prompt-include-backed persistent memory workflow, standard ~/memory files, and the memory-skill contract.
agent/prompt-and-execution.md|Prompt And Execution|Prompt assembly order, message markers, execution transcript rules, and compaction behavior.
agent/skills-and-documentation.md|Skills And Documentation|Skill discovery rules, top-level versus nested skills, conflict rules, and the documentation skill/helper contract.
server/request-flow-and-pages.md|Request Flow And Pages|Exact server routing order, page shell contracts, auth gating, and direct app-file fetches.
server/jobs-and-maintenance.md|Jobs And Maintenance|Primary-owned periodic jobs, interval scheduling, mutation publishing, and guest cleanup maintenance rules.
server/api/files.md|File APIs|The authenticated file endpoints, path forms, writable discovery, folder downloads, and optional local history APIs.
server/api/modules-and-runtime.md|Module And Runtime APIs|Module endpoints, login/runtime endpoints, extensions_load, and identity helpers.
server/auth-and-sessions.md|Auth And Sessions|User storage layout, sealed password/session records, and login/runtime auth behavior.
server/customware-layers-and-paths.md|Customware Layers And Paths|Logical-versus-disk paths, permission rules, optional writable-layer history, override order, and maxLayer.
cli/commands-and-runtime-params.md|Commands And Runtime Params|space.js, command families, runtime-param precedence, and the current schema surface.
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 · 77 lines · 8 tokens per session scan A 355c418d4a14
Documentation is a skill published in the GitHub repository binary16labs/prime-silo (5 stars, last pushed 9d ago), licensed MIT. It adds 8 tokens to every session and 1,386 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it A with 0 findings. It is 92% identical to Documentation, differing in 13 lines, and is treated as a copy.
Other skills, from other repositories
draw-image
Generate an image from a text prompt using an OpenAI-compatible image generation API (gpt-image-1-mini or compatible). The image is uploaded to the gofile.io public file sharing service and ONLY the public download page URL is returned. Trigger when user asks to draw, paint, generate, or create an image.
compose:subagent
Use when executing implementation plans with independent tasks in the current session.
compose:worktree
Use when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback.
compose:merge
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup.
compose:new-skill
Use when creating new skills or editing existing skills for the project or personal skill library.
compose:plan
Use when you have a spec or requirements for a multi-step task, before touching code.