Borrowing it
Nothing to install: this file belongs to rhinocap/raven-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/rhinocap/raven-mcp/main/.claude/skills/grab-falsify-loop/SKILL.mdgit clone --depth 1 https://github.com/rhinocap/raven-mcpWrote 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/rhinocap/raven-mcp/grab-falsify-loop)<a href="https://agentmods.dev/skills/rhinocap/raven-mcp/grab-falsify-loop"><img src="https://agentmods.dev/badge/skills/rhinocap/raven-mcp/grab-falsify-loop/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/rhinocap/raven-mcp/grab-falsify-loop"><img src="https://agentmods.dev/badge/skills/rhinocap/raven-mcp/grab-falsify-loop.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 | $0.00074 | $0.01276 |
| Opus 5 | $0.00037 | $0.00638 |
| Sonnet 5 | $0.00015 | $0.00255 |
| Haiku 4.5 | $0.00007 | $0.00128 |
Grade A, and why
grab-falsify-loop 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 10d 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 — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Grab overlay falsify loop
Every substantive change to browser/raven-grab.js runs this loop until a falsify wave returns only MINORs or accepted-by-design findings.
Invariants (violations are bugs)
- Mirror:
browser/raven-grab.jsmust stay byte-identical toweb/public/raven-grab.js. After every edit:cp browser/raven-grab.js web/public/raven-grab.js && cmp browser/raven-grab.js web/public/raven-grab.js. - Style contract (overlay only):
var+functionkeyword; NO arrow functions, NO optional chaining, NOconst/letinside function bodies, NOfor...of. Template literals: 23 pre-existing backticks are grandfathered — match the surrounding string-concat idiom, don't add new ones. Nested function declarations ARE used (e.g.commit()inbeginStyleEdit). Modern ESM/arrows are fine intest/grab-bridge.test.mjs. - Tests:
npm testmust exit 0 (runs tsc build +node --test test/**/*.test.mjs). The suite honorsRAVEN_GRAB_TEST_OVERLAY=<path>to load an alternate overlay copy. - isConnected convention: the fake test DOM may leave
isConnectedundefined — check=== false/!== false, never truthiness. - buildLayerTree caps: 500 nodes / depth 12.
The loop
- Fix/feature leg (codex,
agentType: 'codex:codex-rescue'): line-anchored spec per edit; edits ONLY the overlay + mirror. Verify withnode --check+ targetednode --test. - Test leg (codex, edits ONLY
test/grab-bridge.test.mjs): one regression test per behavior. Kill-when-reverted proof is mandatory for fixes: revert the specific fix in a scratch copy, run the test against it viaRAVEN_GRAB_TEST_OVERLAY, show it FAILS; then show it passes against the real overlay. A test that can't be killed proves nothing. - Falsify panel (2–3 codex report-only lenses in parallel): prompts start with
REPORT ONLY. DO NOT EDIT ANY FILES.Escalate wave-over-wave: single actions → paired/compound gestures → async interleavings (1.5s layer poll, fetch resolutions, setTimeout timers, deactivate/reactivate). Require severity + exact lines + concrete failure scenario + whether an existing test catches it; tell them an empty blocking list is acceptable. - Main-loop disposition: verify EVERY finding against source before accepting — falsifiers over-state (e.g. the template-literal "contract blocker" that was pre-existing convention). Classify: fix now / accepted-by-design (log why) / deferred (log where).
- Converge: only MINORs/accepted left →
cmp+npm test+ eyes-on harness pass → commit with explicit pathspec.
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.
- 10d ago First seen · 42 lines · 74 tokens per session scan A 2710703974f2
grab-falsify-loop is a skill published in the GitHub repository rhinocap/raven-mcp (6 stars, last pushed 17d ago), licensed Apache-2.0. It adds 74 tokens to every session and 1,276 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 2026-08-31.
Other skills, from other repositories
browse
Drive a real browser through Aside: open a page, read it, click through a flow, take screenshots, check console errors. (gstack).
playwright-cli
A command-line tool for controlling Chromium, Firefox, and WebKit browsers, including navigation, page interaction, screenshots, PDFs, and recorded actions.
webapp-testing
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
create-verification-skill
Create a repo-local verification skill and exhaustive feature map for driving a real app through its UI, CLI, or API. Use for "create a verification skill", "make a verify skill for this repo", or "document how agents can verify this app".
browser-qa
A browser-based quality check for deployed web pages and user flows. It uses browser automation to test rendering, navigation, forms, interactions, responsive behaviour, and accessibility-related issues.
test-electron-app
Drive the real running PostHog Electron app (live tRPC, workspace-server, real data) over CDP with agent-browser. Connect to the running app on port 9222, test desktop changes against a local Django stack, snapshot the accessibility tree, inspect network requests, and screenshot only when explicitly asked. Use when…