Adds a one-command shareable HTML mode to /last30days. The skill detects
HTML intent (explicit --emit=html / --emit:html / --html flag in
$ARGUMENTS, or natural-language asks like "give me a shareable brief",
"for Slack", "export as HTML"), runs the normal research + chat synthesis
flow, then saves a self-contained HTML file to
~/Documents/Last30Days/{topic}-brief.html. The synthesis appears in chat
as usual; the HTML is an additional artifact for sharing.
User experience:
/last30days OpenClaw --emit=html
/last30days OpenClaw, give me an HTML brief for Slack
Synthesis prints to chat. Last line of the response: "📎 Shareable brief
saved to ~/Documents/Last30Days/openclaw-brief.html". Open it, drag it
into a message, browser-print to PDF, email it.
Architecture:
- SKILL.md gets a small detection block (triggers + early exit +
MUST/MUST NOT rules + rationale) that points to a reference file.
- references/save-html-brief.md owns the implementation: capture the
synthesis verbatim into a temp file via heredoc, invoke the engine
with --emit=html --synthesis-file, save to disk, append the
confirmation line to chat.
- lib/render.py exposes render_for_html(report, synthesis_md=None) and
render_for_html_comparison(...) -- clean markdown for HTML
conversion. Omits debug file header, model-facing safety note, and
data quality warnings (those stay in engine stderr; recipients can't
act on them in a shared artifact).
- lib/html_render.py is a new module: ~200-line CSS template (dark
mode default, prefers-color-scheme switch, print stylesheet, mobile
breakpoint), stdlib-regex markdown-to-HTML converter, marker-based
META + engine-footer wrapping, PROSE_LABELS registry promoting plain
-text labels to <h2>, colophon builder.
- last30days.py adds --emit=html argparse choice and --synthesis-file
PATH flag (engine still callable directly without the skill in the
loop).
Design:
- Voice-led research brief, not corporate report. Inter + JetBrains
Mono via Google Fonts with full system fallbacks (no FOIT, works
offline). Brand purple #a855f7 (#7c3aed in light mode). Type ramp:
body 17px/400/muted, bold lead-in 17px/600/fg, h2 + .prose-label
20px/600/fg, monospace badge/meta/footer/colophon at 13-13.5px.
- 720px max-width, generous whitespace, no card layouts or shadows.
- Print stylesheet: light theme, A4 margins, [href]::after URL
footnotes, page-break-inside:avoid on the engine footer.
Templated (locked) shell:
- HTML5 boilerplate, Google Fonts <link> with preconnect, all CSS
inline.
- .badge / .meta / .engine-footer / .colophon containers.
Flexible (role-based):
- <h2> rendering covers BOTH plain ## headers (comparison mode per
LAW 4 exception) AND promoted prose labels via PROSE_LABELS
registry. Adding a new SKILL.md prose label is a one-line tuple
addition; no CSS or template changes.
- Marker-based engine boundaries (<!-- META: ... -->,
<!-- PASS-THROUGH FOOTER -->) survive the markdown converter and
get promoted post-conversion. Robust to engine output format
changes.
- Generic markdown-to-HTML for body content; future SKILL.md additions
(new sections, tables, blockquotes) render correctly without code
changes.
Tests: 30 new tests in tests/test_html_render.py covering snapshots
(rich/thin/comparison), CLI parsing, --synthesis-file end-to-end, prose
label promotion, warning exclusion from artifact, parseability via
html.parser, no-script self-containment.
No SKILL.md voice contract changes, no LAWs 1-8 changes, no new pip
dependencies, no JavaScript anywhere.
Three independent Opus 4.7 self-debugs on 2026-04-18 converged on the same
root cause of the v3.0.6/v3.0.7 canonical-compliance regression: SKILL.md is
42,860 tokens / 1,478 lines, LAWs lived at line 1094+, every realistic reading
strategy failed to reach them before synthesis.
Unit 1 - Moved the BADGE MANDATORY block and VOICE CONTRACT LAW 1-5 (plus
the formatting-authority preface) from line ~1090 to line ~75 (right after
the SKILL CONTRACT preface, before HOW TO INVOKE THIS SKILL). Every reading
strategy now lands the LAWs in active context before synthesis.
Unit 2 - Engine now emits the badge as the first line of --emit=compact
stdout. Passing through the script output becomes the default-correct
behavior; emitting the badge no longer depends on model compliance. Reads
version from .claude-plugin/plugin.json at runtime with graceful fallback.
Unit 3 - Deleted skills/last30days/SKILL.md stub (231-line v3-spec file).
This was the wrong-file-capture hazard Ron Conway's self-debug identified:
model grabbed the first SKILL.md find surfaced and treated it as
authoritative. Only ONE SKILL.md in the plugin package now.
Diagnoses verbatim:
- Kanye thread: "I read lines 1-600 in chunks, jumped to 300-899, then
stopped. File is 1478 lines. I never saw past ~900."
- Peter thread: "I tried Read once, hit the 25K token cap on a 42,860-token
file, and bailed instead of chunked-reading with offset/limit. I never
opened SKILL.md at all."
- Ron Conway thread: "I read one SKILL.md (231 lines)... the v3 spec stub.
I never opened the operational SKILL.md sitting next to the script."
Validation: direct engine invocation confirms badge at line 1 of compact
output. Module imports clean.
Co-authored-by: Matt Van Horn <455140+mvanhorn@users.noreply.github.com>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Atomic bump across all four manifests:
- SKILL.md (root)
- skills/last30days/SKILL.md (internal spec)
- .claude-plugin/plugin.json
- gemini-extension.json
CHANGELOG entry documents the skill-upload packaging fix, vendor/ removal,
legacy plans/ removal, and the new scripts/build-skill.sh builder.
Two SKILL.md files declared `name: last30days` with `user-invocable: true`,
which caused strict marketplace validators to reject the plugin with "Some
plugins in this marketplace have validation errors":
- ./SKILL.md (canonical, also reachable via skills/last30days-nux/ symlink)
- ./skills/last30days/SKILL.md (v3 architecture spec, real file)
In v2.9.6, skills/last30days/SKILL.md was a symlink to ../../SKILL.md so
only one skill existed. Commit 0a9ff16 (v3.0.0) added a new real file at
skills/last30days-v3/SKILL.md, and commit 9be0780 then renamed that
directory to skills/last30days/, replacing the original symlink with a
different real file. The collision has been live since v3.0.0 shipped.
This change:
- Renames skills/last30days/SKILL.md to name: last30days-v3-spec and sets
user-invocable: false. The file stays in place as internal architecture
documentation, but it no longer competes with the canonical skill.
- Fixes README.md link that pointed to the deleted skills/last30days-v3/
path (left over from the rename).
- Removes a stale variants/open/SKILL.md reference (variants/open was
deleted in v3.0.0).
After the change, only one canonical name=last30days user-invocable=true
skill exists (the root SKILL.md, also reachable via the
skills/last30days-nux/ symlink, same inode).
Closes#204.
This contribution was developed with AI assistance (Codex).
Co-authored-by: Matt Van Horn <455140+mvanhorn@users.noreply.github.com>
Add gemini-extension.json manifest with correct array-format settings,
symlink skills/last30days/SKILL.md to root SKILL.md for Gemini skill
discovery, add Gemini install paths to bash for-loop in both main and
open variant, and add Gemini CLI install instructions to README.
Incorporates the good parts of PR #53 (manifest, paths, README) while
avoiding duplicate SKILL.md, tool name scattering, and allowed-tools
pollution that would have created maintenance issues.
Closes#45
Co-Authored-By: Alex Ferrari <alex@thealexferrari.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Trevin Chow