Gno

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

If approved too broadly, the agent could change the local Gno index, tags, context, or integrations while performing document-search tasks.

Why it was flagged

The agent can invoke any `gno` subcommand, not just read-only searches. The documented command set includes indexing, tagging, cleanup/reset, MCP setup, publishing export, and skill installation, so users should treat Gno CLI calls as potentially mutating even though this is disclosed and aligned with the skill’s purpose.

Skill content
allowed-tools: Bash(gno:*) Read
Recommendation

Use the skill for specific user-requested Gno tasks, review commands that install integrations, reset/cleanup data, modify tags/context, or publish/export content, and avoid blanket approval of high-impact commands.

What this means

Installing or updating the external Gno CLI gives that software local execution access on the host.

Why it was flagged

The skill depends on an externally installed CLI, including a remote shell installer and global package installation. This is normal for a CLI-driven skill, but the executable behavior is outside the provided artifact scan.

Skill content
curl -fsSL https://gno.sh/install | bash

# npm / Bun
bun add -g @gmickel/gno        # or: npm install -g @gmickel/gno
Recommendation

Install Gno only from trusted upstream sources, verify the package or installer before running it, and keep the CLI updated from the same trusted channel.

What this means

Sensitive local documents may become searchable by the agent, and stored context hints could bias future search results or answers.

Why it was flagged

Gno can index broad file patterns and store context hints that influence later search relevance. This is central to a local knowledge engine, but it means private documents and persistent retrieval hints can be reused across later tasks.

Skill content
`--pattern <glob>` File pattern (default: `**/*`) ... `--exclude <paths>` Exclude patterns (default: `.git,node_modules`) ... `gno context add <scope> "<text>"`
Recommendation

Index only intended folders, use include/exclude patterns for secrets and private data, periodically review collections/context, and remove or reset indexes that should no longer be available.

NoteHigh Confidence
ASI10: Rogue Agents
What this means

After MCP installation, future AI sessions may continue to have Gno access until the MCP configuration is removed, and write-enabled mode may allow modifications through the MCP interface.

Why it was flagged

The skill documents installing Gno as an MCP server in AI client configurations, optionally with write tools. This is disclosed integration behavior, but it creates a persistent path for AI clients to access or modify Gno-managed data.

Skill content
gno mcp install

gno mcp install -t claude-code

# With write tools enabled
gno mcp install --enable-write
Recommendation

Install MCP only when needed, avoid `--enable-write` unless you specifically require it, check status after installation, and uninstall the MCP server when persistent access is no longer desired.