sync.sh was written against the layout of mvanhorn/last30days-skill-private
(`.../cache/last30days-skill-private/last30days-3/{version}`) and that path
was never updated when this public repo got its own copy. Running sync.sh
from here populated the BETA channel's cache (`/last30days-beta`) instead
of this repo's own `/last30days` cache, so devs working in this repo could
not test their changes via the public slash command without waiting for a
marketplace release.
Path now derives from this repo's own manifests:
- marketplace name `last30days-skill` (.claude-plugin/marketplace.json)
- plugin name `last30days` (.claude-plugin/plugin.json)
Drops the `last30days-3-nogem` target along with it - that's a private-repo
variant with no public equivalent.
Updates test_sync_cache_path_uses_skill_version to assert the new path
pattern and clarifies the COMMON_TARGETS comment so the next person editing
it understands which marketplace/plugin name segments come from where.
Five provider modules (pinterest, threads, instagram, tiktok, youtube_yt)
and watchlist.py each carried a try/except `requests` import with parallel
urllib + requests branches. The urllib path already used the
stdlib-only wrapper at `lib/http.py` (retries, 429 handling, HTTPError).
This collapses every dual-branch into a single `http.get`/`http.post`
call and removes the `requests` dependency from `pyproject.toml`.
Also drops 4 transitive deps (urllib3, certifi, charset-normalizer, idna)
from the lockfile, leaving the skill stdlib-only at runtime.
Tests for tiktok comments and watchlist delivery were rewritten to mock
`lib.http` directly instead of the now-removed `requests` module.
Out of scope but flagged during review: the 13 surviving SC call sites
share a near-identical scaffold and would benefit from a
`http.scrapecreators_get(url, params, token, ...)` helper. Filed for a
follow-up PR rather than expanding scope here.
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.
`tests/test_plugin_contract.py::test_versions_match_across_manifests`
enforces that every version-bearing surface agrees: pyproject.toml,
SKILL.md, both plugin.json files, AND the marketplace plugin entry.
The Claude Code spec says plugin.json wins when both are set, but this
repo deliberately mirrors the version across all surfaces and tests it.
Restore the field at 3.1.1 to satisfy the contract.
Codex CLI reads AGENTS.md for repo-level context the way Claude Code reads CLAUDE.md. Delegate to the existing CLAUDE.md so both harnesses share one source of project instructions.
- Remove no-op `"hooks": {}` from .claude-plugin/plugin.json (auto-discovery from hooks/hooks.json picks up the SessionStart hook).
- Remove redundant `version` from marketplace.json plugin entry; plugin.json is the source of truth per the spec.
- Sync description / longDescription across .claude-plugin and .codex-plugin manifests so all surfaces show the same copy.
marketplace.json had stale v3 rename artifacts: plugin name was
"last30days-3" (should be "last30days" to match plugin.json) and
version was "3.0.0-alpha" (should be "3.0.0" to match the stable
release).
The module-level _cached_token was set once and never refreshed. AT
Protocol tokens expire after ~2 hours, causing silent 401 errors in
long-running watchlist cron sessions. Adds a 90-minute expiry check
using time.monotonic() and logs re-authentication.
Fixes#92
Add column whitelists to prevent SQL injection via kwargs keys in
dynamic UPDATE queries. Values were already parameterized but column
names were string-interpolated directly from kwargs.
Fixes#90
gemini-extension.json still referenced v2.9.5 and
skills/last30days-v3/SKILL.md still said 3.0.0-alpha.
Both now match pyproject.toml's canonical 3.0.0 version.
Addresses items from #190. Structural decisions (SKILL.md
consolidation, SKILL-original.md cleanup) left for maintainer.
This contribution was developed with AI assistance (Claude Code).
The second assertion in test_resolve_google_judge_api_key_prefers_google_key
ran outside the mock.patch.dict context. When GOOGLE_API_KEY or GEMINI_API_KEY
is set in the real environment, os.environ takes precedence over the config
dict fallback and the test fails.
Wrap the assertion in its own mock.patch.dict scope that clears the three
relevant env vars so the test passes regardless of the developer's env.
This contribution was developed with AI assistance (Claude Code).
lib/exa_search.py was removed during the v3 refactor but
tests/test_exa_search.py still imports from it. This causes
an ImportError that blocks pytest -x from running any tests.
This contribution was developed with AI assistance (Claude Code).
When Bird's JSON response is a raw array instead of an object,
json.loads returns a list. All callers use .get('items') which raises
AttributeError on lists. Wrap list responses in {"items": parsed} so
callers always receive a dict.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>