Sec Daily Digest
PassAudited by ClawScan on May 10, 2026.
Overview
The skill’s behavior matches its stated digest-generation purpose, but it uses expected API keys, external AI/Twitter/email services, and local persistent state that users should review before enabling.
Before installing, decide which integrations you actually need: dry-run avoids AI calls, --no-twitter avoids Twitter/X, and omitting --email avoids Gmail sending. If you enable AI or Twitter/X, provide only the needed API keys and remember that article excerpts may be sent to the selected provider. Review ~/.sec-daily-digest periodically if you want to clear stored digest history or source settings.
Findings (5)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A hostile RSS article or web page could bias classifications, summaries, or the digest text, although the artifacts do not show that model output is used to execute tools.
Fetched article descriptions and optional full-text excerpts are inserted directly into AI prompts. This is necessary for summarization/scoring, but a malicious article could try to steer the model’s output.
entry += `\n\n[Full text excerpt]\n${article.fullText.slice(0, 1500)}`; ... `严格输出 JSON`Treat generated summaries as advisory, and consider adding stronger prompt wording that labels article content as untrusted data.
If the user enables --email, the skill can cause an email to be sent through the local gog/Gmail setup to the supplied recipient.
The optional --email feature invokes a local Gmail CLI to send the generated digest. This is user-directed and disclosed, but it can send email from the user’s configured account.
gog gmail send --to <addr> --subject "sec-daily-digest YYYY-MM-DD" --body-file -
Use --email only with an intended recipient and a trusted local gog configuration; otherwise generate the Markdown file without email delivery.
Provider or Twitter/X API keys may be used to call those services and may incur usage or expose fetched article content to the chosen AI provider.
The skill discloses expected AI-provider and Twitter/X credentials. These credentials are purpose-aligned, but they grant access to third-party services.
OPENAI_API_KEY ... GEMINI_API_KEY ... ANTHROPIC_API_KEY ... TWITTERAPI_IO_KEY ... X_BEARER_TOKEN
Provide only the credentials needed for the selected provider/features, prefer least-privilege keys where available, and avoid setting optional keys unless you want those integrations.
Installing dependencies may resolve a newer compatible yaml package version than the one originally tested.
The package relies on an npm dependency using a caret version range. This is common, but without a lockfile in the provided manifest it is less reproducible.
"dependencies": { "yaml": "^2.8.2" }Install from a trusted source and use a lockfile or pinned dependency versions if reproducible builds are important.
Digest history, source preferences, and health data remain on disk and influence future deduplication/health warnings.
The skill stores local state across runs for configuration, source health, OPML cache, and article archives. This persistence is disclosed and scoped to the skill’s state directory.
Config root: `~/.sec-daily-digest/` ... Health file: `~/.sec-daily-digest/health.json` ... Archive dir: `~/.sec-daily-digest/archive/`
Review or delete ~/.sec-daily-digest if you want to reset history or remove stored source information.