Turn bs-report-<slug>-<date>.md into one HTML file anybody can open.
The markdown is the artifact and stays the artifact — it is what tally.py checks, what gets\ndiffed against a later run, and what gets committed. This skill adds a reader-facing view of\nthat same file. It renders; it never edits, recounts, or adds. Every number on the page is copied\nfrom the markdown verbatim.
A BS report is a five-column claims table. On a laptop that is fine. On a phone — which is where a\nshared link is actually opened — the Evidence column is unreadable, and the reader's first\nquestion (\"what's wrong with it?\") means scrolling twenty rows to find the two that matter.
\nuv run <this-skill-dir>/scripts/render_report.py ~/.bullshit-detector/reports/<YYYY>/bs-report-<slug>-<date>.md\nWrites bs-report-<slug>-<date>.html beside the markdown. -o <path> puts it elsewhere. No install\nstep, no dependencies — the script is stdlib-only, runs under plain python3 as well as uv run,\nand the page it produces makes no network requests. That combination is what lets it work in a\ncode-execution sandbox, where the HTML is the only thing the user can take away.
--open shows the page in the default browser. Where there is no browser — a sandbox, a headless\nhost — it says so and the file is still written. It is skipped when the page already exists,\nbecause re-rendering is normal (the run line gets finalised after a first pass) and every open\nspawns another tab — three consecutive real runs left the user with two. The file updates in\nplace; reload the tab you have.--reopen opens even when the page existed — for picking a report back up in a later session. Never treat that as a failure.--quiet prints only the output path, for scripting.By default the script prints the block that ends a detector run:
\nBS score 4/10 · Mostly fine\n the macro data is real and mostly checks out; the narrative glue is crypto-Twitter.\nTally: 35 claims extracted, 34 individually source-checked — 22 confirmed, 5 plausible,\n 5 misleading, 2 false. 1 not checked.\nrun: 16m30s, searches 35, tools 65, coverage 1, per claim 29s\n\nmarkdown file:///Users/…/bs-report-japans-money-is-collapsing-2026-07-31.md\npage file:///Users/…/bs-report-japans-money-is-collapsing-2026-07-31.html\n opened in your browser\nPaste it; don't rebuild it. Every figure came out of the report that tally.py had just\nrecounted. A summary retyped from memory of what you wrote is wrong in the direction that flatters\nthe run — the same failure mode as the tally and the search count, one level up.
The paths are file:// URLs on purpose: terminals linkify a bare URL, so both files are one\ncmd-click from opening. Don't shorten them to ~/… or hide them behind link text.
Before rendering, the script runs the report through the detector's tally.py and refuses to\nrender a report that fails it:
REFUSED: bs-report-our-solar-system-2026-07-31.md does not pass tally.py.\n ✗ run line: 25 claims individually source-checked from 21 searches — every claim\n carrying a verdict needs its own search, so this reports more verification than\n was performed\nExit 3. This is the point of the gate: a page this presentable, built from a report that fails its\nown arithmetic, launders a broken report into something that looks authoritative. Fix what the gate\nnames and re-run.
\n--force renders anyway and prints the failures as warnings.--no-check skips the gate — for markdown that isn't a BS report.--tally <path> or $BULLSHIT_DETECTOR_TALLY if it can't be found automatically.If tally.py isn't installed at all — this skill can be installed without the detector — the\nscript warns and renders. A missing validator is a reason to warn, not a reason to stop someone\nviewing a report they already have.
--og-image <absolute-url> adds a link-preview image. Only useful once the page is hosted\nsomewhere; skip it for local files.
For a hosted page the script also emits <link rel=\"canonical\">, Open Graph and Twitter card\ntags, and Article structured data. It needs the page's public URL to do that:
--canonical <absolute-url> is that URL. Pass it whenever the page has a fixed address.url in the structured data. The script never\nguesses one from --og-image or anything else: a wrong canonical tells search engines the wrong\naddress is the real one, which is worse than none.--author <name> and --author-type Person|Organization set the Article author. The default is\nthe tool itself, since this skill runs for everyone; pass your own name when you publish.
datePublished comes from the report's own **Checked:** line, never from the render time. A\nfact-check that stamps today's date on work done last month is the one thing this structured data\nmust not say, so a date it cannot read produces no date at all.
Then tell the user the path and that it opens in any browser. On macOS open <path> does it.
Problems shows only ❌ and \uD83D\uDFE0. Load-bearing only hides the\nincidental table. This is the whole reason a non-technical reader gets through the report.<title> — N/10 and the verdict line instead of a\nbare URL.tally.py validated.--force to make a refusal go away. The gate is naming a real defect in the\nreport. Rendering past it produces a page that looks more trustworthy than the thing behind it,\nwhich is the failure this whole tool exists to catch.tally.py is\nthe strict one, and the two must not swap roles.| # | Claim | Type | Verdict | Evidence | shape — the # column\nis how the renderer tells a claims table from any other table, and the verdict glyph is read from\nits own cell, never from anywhere else in the row.