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/cognitive-fab/polygraph/polyviznpx skills add cognitive-fab/polygraph --skill polyvizgit clone --depth 1 https://github.com/cognitive-fab/polygraphWhat 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 | $0.00206 | $0.01312 |
| Opus 5 | $0.00103 | $0.00656 |
| Sonnet 5 | $0.00041 | $0.00262 |
| Haiku 4.5 | $0.00021 | $0.00131 |
Grade A, and why
polyviz 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 — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
polyviz — visualize the verification, don't redraw it
polyviz is the VISUALIZATION side of the Polygraph method (Polygraph audits,
polygen authors, polyrun executes, polyvers evolves, polynv elicits — polyviz
draws). Every mark it emits traces to a field in a Polygraph artifact, so the
picture cannot drift from the model that was actually checked. Rendering is a
pure function artifacts → SVG: no Date.now(), no randomness, sorted keys —
byte-identical on every run. An LLM may invoke polyviz; it never draws.
The CLI is ${CLAUDE_PLUGIN_ROOT}/bin/polyviz.mjs:
polyviz render --in <dir|polyviz.json> --diagram <all|state-machine|invariants|counterexample|compat-gate|model-card> \
--out <dir> [--format svg[,png]] [--theme dark|light] [--tokens tokens.json] [--scale 2]
polyviz hash --in <dir|polyviz.json> --diagram <...> # sha256 per SVG, no files (determinism/CI check)
polyviz schema # print the viz-model JSON Schema
Two ways to feed it
-
A viz-model JSON (
*.polyviz.json) — the stable contract (polyviz schemaprints it). Pureviz-model → SVG: no user code executed, no optional deps needed for the SVG path. Best when you're authoring figures by hand or a tool already emitted a viz-model. -
A Polygraph/polyvers artifacts directory (
--in <dir>) — adapters map the native artifacts to a viz-model:contract.json+ its SAM/nextmodule → the machine graph. The contract has no edge list, so transitions are derived by a bounded, deterministic BFS that executes the module over its declared(action,data)domain. This runs the target module — only do it on a run you trust.- the invariants module (
invariants.mjs) → the invariants panel. - a
findings.jsonreporting a failure, resolved to its trace → the counterexample. - polyvers
compat-report.json→ the compat-gate. Drop apolyviz.annotations.jsonin the dir to supply the narrative the raw artifacts don't carry (titles, nicer invariant text, version-card labels, a highlighted state).
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 · 82 lines · 0 tokens per session scan A cf4d60f34b32
polyviz is a skill published in the GitHub repository cognitive-fab/polygraph (11 stars, last pushed 6d ago), licensed Apache-2.0. It adds 206 tokens to every session and 1,312 once invoked, about $0.0010 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
adev-writing-guide
Comprehensive writing guide for Angular documentation (adev). Covers Google Technical Writing standards, Angular-specific markdown extensions, code blocks, and components. You MUST use this skill any time you plan to create, edit, or review documentation files in adev/ or adev/src/content.
reference-signal-forms
Explains the mental model and architecture of the code under packages/forms/signals. You MUST use this skill any time you plan to work with code in packages/forms/signals.
lint-js
Lint JS/TS code only. Use before opening a PR when only JavaScript or TypeScript files were changed (no Rust).
compiler-commit
Use when you want to verify compiler changes and commit with the correct convention. Runs tests, lint, and format, then commits with the [compiler] or [rust-compiler] prefix.
reference-compiler-cli
Explains the mental model and architecture of the code under packages/compiler-cli. You MUST use this skill any time you plan to work with code in packages/compiler-cli.
Shade dropdown surface contract
DropdownMenu, Select, and Popover share one visual recipe (bg-surface-elevated-2 + border-border/60 dark:border-border/30 + shadow-md). Change them together. Trigger when editing any of those three Shade files.