Clawdbot documentation expert with decision tree navigation, search scripts, doc fetching, version tracking, and config snippets for all Clawdbot features
Security Analysis
medium confidenceThe skill's requirements and runtime instructions line up with a documentation/navigation helper for Clawdbot; it does not request unexpected credentials or installs, but it does include runnable scripts that will access network and disk so review before use.
Name/description, SKILL.md, snippets, and the included scripts all focus on discovering, fetching, indexing, and surfacing Clawdbot docs; the requested capabilities (sitemap, search, fetch, track-changes) are coherent with the stated purpose.
Runtime instructions direct the agent to run the provided scripts and to fetch docs from docs.clawd.bot; they do not ask the agent to read unrelated system files or request unrelated secrets. The scope stays within documentation navigation and retrieval.
There is no install spec (instruction-only), which minimizes install risk. The skill does reference external tooling (qmd) for full-text indexing, but it does not itself download arbitrary archives or run an installer.
The skill declares no required environment variables, credentials, or config paths. This is proportionate to its documented function. Note: the scripts will perform network fetches and may write snapshots to disk (track-changes), but they do not request secrets in the metadata.
The skill is not marked always:true and requests no special privileges. Model invocation is not disabled (default allows the model to call the skill autonomously), which is typical for helper skills but worth noting if you want explicit user confirmation before running scripts.
Guidance
This skill appears to be a coherent documentation helper for Clawdbot and doesn't ask for keys or installs. However, it includes small shell scripts that will fetch docs from the network and create local snapshots/indexes. Before installing or enabling the skill: 1) Inspect the contents of the scripts (build-index.sh, fetch-doc.sh, track-changes.sh, etc.) to confirm they only access docs.clawd.bot and do not exfiltrate local files. 2) Run the scripts in a sandbox or with network controls if you are concerned about network writes. 3) Check package.json for any unexpected dependencies. 4) If you prefer to prevent autonomous actions, disable model invocation for this skill or require explicit user consent before the agent runs any scripts.
Latest Release
v1.2.2
Added YAML frontmatter description for ClawdHub summary display
Popular Skills
Published by @NicholasSpisak on ClawHub
