brain-pdf

brain-pdf — Render a Brain Page to Publication-Quality PDF

Convention: see conventions/quality.md (via get_skill conventions/quality) for output rules. The PDF is a rendering — never the primary artifact. If a PDF exists, the source brain page exists behind it.

The rule

The brain page is ALWAYS the source of truth. The PDF is a rendering of it, never a standalone artifact. If a PDF exists somewhere, the brain page must exist behind it.

What this does

Renders a brain page (markdown with frontmatter) into a publication-quality PDF using the host agent's PDF renderer (e.g. the agent harness's make-pdf skill or an equivalent markdown-to-PDF toolchain). Output is suitable for:

• Sharing a personalized book mirror via email or chat
• Delivering a strategic-reading playbook as a clean read
• Producing a briefing or report with running headers and page numbers
• Archiving a long-form essay in a portable format

Prerequisite: a host-side PDF renderer

This skill depends on a markdown-to-PDF renderer available on the host where the agent runs. memex itself does not render PDFs — it serves the page body; rendering is the agent's job.

Verify a renderer is available before invoking (check for the agent harness's make-pdf skill, or a pandoc/Chromium-print pipeline). If none is installed, the skill cannot run — tell the user which prerequisite is missing instead of improvising a degraded rendering.

Workflow

1. RESOLVE  → Confirm the brain page exists (page_get <slug>).
2. STRIP    → Remove YAML frontmatter — the renderer would otherwise
              dump it as a full page of raw metadata text.
3. RENDER   → Invoke the PDF renderer with sane defaults (no cover,
              no TOC).
4. DELIVER  → Hand the PDF to the requester via the agent's preferred
              file-delivery channel (an explicit attachment tool — not
              inline media markup, which can fail silently).

Invocation

SLUG="path/to/page"

# 1. Confirm the page exists and pull the raw markdown body.
#    memex is DB-canonical: there is no repo checkout of the page —
#    the body always comes from the brain over MCP.
RAW=$(mktemp "$TMPDIR/brain-page-XXXXXX.md")
memex call page_get "{\"slug\":\"$SLUG\"}" > /dev/null \
  || { echo "Page $SLUG not found" >&2; exit 1; }
memex call page_get "{\"slug\":\"$SLUG\"}" | jq -r '.content' > "$RAW"

# 2. Strip YAML frontmatter — sed: skip the opening '---' through the
#    closing '---' (lines 1..N), then keep everything after.
CLEAN=$(mktemp "$TMPDIR/brain-page-clean-XXXXXX.md")
sed '1{/^---$/!q}; /^---$/,/^---$/d' "$RAW" > "$CLEAN"

# 3. Render. NO cover, NO TOC by default — they look corporate
#    and waste space. Add them only if explicitly requested.
OUT="$TMPDIR/$(basename "$SLUG").pdf"
# invoke the host renderer, e.g. the agent's make-pdf skill:
#   make-pdf "$CLEAN" "$OUT"

echo "Rendered: $OUT"