Fund News Summary

SuspiciousAudited by ClawScan on May 10, 2026.

Overview

The skill gathers market news, but it also schedules automatic runs and can send reports to Telegram and push them to a hard-coded GitHub repository, so it should be reviewed before use.

Install only if you intentionally want a daily automated fund-news workflow. Before using it, verify the cron job, Telegram destination, Obsidian path, GitHub repository/branch, credentials, and qveris dependency; disable automatic sending or pushing unless you have reviewed and approved those destinations.

Findings (6)

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.

ConcernHigh Confidence
ASI10: Rogue Agents
What this means

The skill may keep running every day and produce/send reports even when the user did not just ask for one.

Why it was flagged

The skill declares recurring background execution, but the artifacts do not show clear opt-in, disable, or per-run review controls.

Skill content
This skill is bound to the `DailyFundNews` cron job, executing automatically at **11:00 Beijing Time** daily.
Recommendation

Require explicit user opt-in for the cron job, document how to disable it, and avoid external sends or pushes during scheduled runs unless separately approved.

What this means

Generated reports could be posted to the wrong Telegram destination or committed to a GitHub repo without the user reviewing them first.

Why it was flagged

The instructions call for messaging and repository mutation, but do not specify approval, destination verification, dry-run behavior, or rollback.

Skill content
Send to Telegram: Send the script output directly to Telegram ... Auto GitHub push: Automatically pushes to GitHub after generation
Recommendation

Show the exact Telegram recipient and GitHub repo/branch before sending or pushing, and require confirmation for each external write.

ConcernMedium Confidence
ASI03: Identity and Privilege Abuse
What this means

It may use whatever GitHub credentials are already present in the environment, and users may not realize which account or repo will be affected.

Why it was flagged

The skill expects authenticated GitHub access to push to a hard-coded repository, while the registry metadata declares no primary credential or required environment variables.

Skill content
GitHub push fails - Cause: Network issue or authentication expired ... GitHub repo: `https://github.com/YonghaoZhao722/yonghao-notes`
Recommendation

Declare the required credentials and scopes, use a dedicated least-privilege token, and make the target repository configurable by the user.

What this means

Installing or running this skill may execute additional code from another local skill or directory that the user has not reviewed.

Why it was flagged

The script invokes an external qveris helper that is not included in this skill's manifest or install spec, so its provenance and behavior are not reviewable here.

Skill content
QVERIS_PATH = os.path.join(os.path.dirname(__file__), "../qveris") ... cmd = ["python3", "scripts/qveris_tool.py", "execute", tool_id, ...]
Recommendation

Declare qveris as a dependency, pin or include the required helper, and document exactly what external tool permissions it needs.

What this means

Reports can reveal the user's market interests or watchlist and may be delivered outside the intended audience.

Why it was flagged

The skill sends generated fund/news reports to an unspecified Telegram destination, leaving the external data boundary unclear.

Skill content
Send to Telegram: Send the script output directly to Telegram
Recommendation

Require the Telegram chat/channel to be explicitly configured and displayed before sending, and avoid automatic delivery unless enabled by the user.

What this means

Old or inaccurate generated summaries could remain in the user's notes and influence later research.

Why it was flagged

The skill persistently stores generated summaries in an Obsidian vault, which is purpose-aligned but may later be reused, synced, or trusted as context.

Skill content
Obsidian save path: `/root/clawd/obsidian-vault/reports/fund/YYYY-MM-DD.md`
Recommendation

Keep the vault location configurable, label generated reports clearly, and review them before syncing or reusing them as trusted notes.