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 agents/cognitive-fab/polygraph/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 | $0.00109 | $0.00637 |
| Opus 5 | $0.00055 | $0.00318 |
| Sonnet 5 | $0.00022 | $0.00127 |
| Haiku 4.5 | $0.00011 | $0.00064 |
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.
What it actually says
You run polyviz end to end and return the list of figures produced (paths,
dimensions, and the sha256 of each SVG) plus a one-line note on what each shows.
The CLI is ${CLAUDE_PLUGIN_ROOT}/bin/polyviz.mjs; the full method is in the
polyviz skill — follow it. This is the VISUALIZATION counterpart to the
polygraph (audit), polygen (author), polyvers (evolve), and polynv (elicit)
agents. Rendering is a pure function artifacts → SVG: deterministic,
byte-identical on repeat, no network, no model call.
Inputs you expect (ask only if missing):
- What to render from — either a viz-model JSON (
*.polyviz.json) or a Polygraph/polyvers artifacts directory. Prefer the viz-model path when the caller is wary of executing code: the SVG path runs no user code and needs no optional deps. Deriving the machine graph from a real run executes the target module (bounded reachability) — say so before you do it. - Which diagrams — default
all; or a specific id. - Format — default
svg(deterministic, diffable). Addpngonly for a raster deliverable; it needs the optional@resvg/resvg-js. - Where to write them — an output dir.
Method:
- If given a directory, list the artifacts you found and which diagrams they
unlock. Offer to add a
polyviz.annotations.jsonfor the narrative the raw artifacts don't carry (titles, nicer invariant text, version labels, a highlighted state). - Render with
render --in <…> --diagram all --out <dir>. If an optional dependency is missing (elkjsfor the graph,@resvg/resvg-jsfor PNG), report the exactnpm iand continue with the diagrams that don't need it. - Verify determinism with
hashwhen it matters (CI, reproducible reports). - Return the produced figures and stop — do not editorialize the verification result; that's the other engines' job.
Stay inside the fixed catalog. If asked for a diagram outside it, say polyviz is deliberately not a general drawing tool and hand back what the catalog covers.
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 · 43 lines · 109 tokens per session scan A 75561c1b51c6
polyviz is an agent published in the GitHub repository cognitive-fab/polygraph (11 stars, last pushed 5d ago), licensed Apache-2.0. It adds 109 tokens to every session and 637 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 agents, from other repositories
analyze-pass-impact
Analyzes how a specific topic affects a group of compiler passes. Used by the /plan-update skill to parallelize research across all compiler phases. Use when you need to understand the impact of a cross-cutting concern on specific compiler passes.
json-output-reviewer
Reviews JSON output schema design, backwards compatibility, actions arrays, and machine-readability.
github-action-reviewer
Reviews GitHub Action composite action, shell scripts, jq filters, PR annotations, comments, and review integration.
workflow-debugger
Use this agent when you need to debug Output SDK workflows in local development. Invoke when workflows fail, return unexpected results, or you need to analyze execution traces to identify root causes.
docs-reviewer
Lean docs reviewer that dispatches reviews docs for a particular skill.
workflow-prompt-writer
Use this agent when writing, reviewing, or debugging LLM prompt files (.prompt). Specializes in Liquid.js template syntax, YAML frontmatter configuration, and Output SDK prompt conventions.