Clawd Docs V2

AdvisoryAudited by Static analysis on Apr 30, 2026.

Overview

No suspicious patterns detected.

Findings (0)

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.

What this means

The agent may use local command-style reads and make network requests to the ClawdBot docs site when answering documentation questions.

Why it was flagged

The skill instructs the agent to read scoped local documentation files and fetch documentation pages from a specific docs domain. This is expected for the stated documentation-access purpose and is not shown performing writes or destructive actions.

Skill content
cat ~/clawd/data/docs-snippets/telegram-setup.md ... web_fetch({ url: "https://docs.clawd.bot/{path}", extractMode: "markdown" })
Recommendation

Use the skill for documentation lookups, and verify important answers against the live docs page when accuracy matters.

What this means

Answers may reflect cached documentation rather than the latest live documentation if the local cache is outdated or tampered with.

Why it was flagged

The skill relies on persistent local cached documentation and an index that may be reused across requests. If those local files become stale or are modified by an untrusted party, the agent could answer from inaccurate cached context.

Skill content
~/clawd/data/ ├── docs-index.json # Search index ├── docs-stats.json # Usage tracking ├── docs-snippets/ # Cached Golden Snippets └── docs-cache/ # Full page cache
Recommendation

Keep the cache directory trusted, refresh from `https://docs.clawd.bot/` for important operational steps, and avoid relying on cached snippets if they look unexpected.