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 skills add mrstev3n/balise-skills --skill balise-content-testgit clone --depth 1 https://github.com/mrstev3n/balise-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/mrstev3n/balise-skills/balise-content-test)<a href="https://agentmods.dev/skills/mrstev3n/balise-skills/balise-content-test"><img src="https://agentmods.dev/badge/skills/mrstev3n/balise-skills/balise-content-test/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mrstev3n/balise-skills/balise-content-test"><img src="https://agentmods.dev/badge/skills/mrstev3n/balise-skills/balise-content-test.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00078 | $0.02668 |
| Opus 5 | $0.00039 | $0.01334 |
| Sonnet 5 | $0.00016 | $0.00534 |
| Haiku 4.5 | $0.00008 | $0.00267 |
Grade A, and why
balise-content-test 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 12d 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 — 234 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Balise Content Test
Work as a content resilience tester inside the current Figma file. Replace ideal placeholder content with difficult but plausible content, observe what fails, propose the smallest durable correction, and preserve a visible trace of the test.
Treat content as any variable data rendered by the interface: text, names, identifiers, numbers, dates, currencies, units, statuses, missing or partial values, user-generated input, localized strings, and externally supplied media.
Protect the Source
- Treat the selected layers, frames, component sets, or section as the working boundary.
- If neither the selection nor the prompt identifies a testable scope, ask the user to select one.
- Inspect parents, nearby states, related components, properties, variables, annotations, and layout conventions when useful. Do not edit them unless authorized.
- Never stress-test directly on source frames. Create clearly labeled duplicates.
- If duplication or editing is unavailable, perform a read-only audit and state the limitation.
- Preserve components, instances, variants, properties, bindings, styles, constraints, guides, and Auto Layout whenever possible.
- Do not detach instances, flatten layers, replace design-system components, or modify shared library assets without explicit authorization.
- Use synthetic data only. Never insert personal, confidential, financial, health, authentication, research, or client data.
- Do not claim production behavior, accessibility compliance, translation quality, or user comprehension from a static Figma test.
Determine the Mode
- Audit: inspect and report risks without changing the canvas.
- Test: create labeled test copies and diagnose them. Use this by default for “stress-test”.
- Repair: repair confirmed failures only in the authorized selection.
- Test and repair: generate cases, diagnose, repair the copies, then re-test.
When edit intent is unclear, default to Audit.
Follow the Evidence Workflow
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.
- 12d ago First seen · 234 lines · 78 tokens per session scan A bfad0829d963
balise-content-test is a skill published in the GitHub repository mrstev3n/balise-skills (11 stars, last pushed 23d ago), licensed Apache-2.0. It adds 78 tokens to every session and 2,668 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
go-testing
Trigger: Go tests, go test coverage, Bubbletea teatest, golden files. Apply focused Go testing patterns.
ui-testing
Visual testing - catch invisible buttons, broken layouts, contrast.
visual-qa
Runs rigorous visual QA across web, terminal, and paginated surfaces with screenshot evidence and a verdict. Use for any UI build or change, or when asked whether a page, component, or TUI looks right.
visual-validation
Autonomous screenshot verification — captures, compares, and AI-evaluates UI changes for visual regressions.
app-user-story-qa
End-to-end app feature inventory and user-story testing workflow with a canonical tracker. Use when the user asks to audit every feature, derive expected behavior from code, test user journeys, or explicitly fix and retest documented UX or logistical defects.
coding-agents-hooks-authoring
To author, register, and test Rosetta hooks, add a SemanticKind, or debug a hook that won't fire.