CtrlK
BlogDocsLog inGet started
Tessl Logo

jbaruch/nanoclaw-untrusted

Security rules for untrusted NanoClaw groups. Credential protection, internal file protection, social engineering defenses.

76

Quality

95%

Does it follow best practices?

Impact

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

Quality

Content

87%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A tight, well-organized etiquette skill with concrete behavioral directives and no token waste. It scores high on conciseness, actionability, and organization; workflow clarity is capped at 2 because the skill is a single-context behavioral policy with no multi-step or validation-bearing process.

DimensionReasoningScore

Conciseness

Lean body that assumes Claude's competence; each line is actionable guidance with no padding or explanation of basic concepts Claude already knows.

3 / 3

Actionability

Gives concrete, copy-ready directives such as 'say you're an AI assistant', 'The filesystem is read-only. Don't attempt writes', and a ready phrase 'I can't do that in this chat'.

3 / 3

Workflow Clarity

Guidelines are clear but this is a behavioral policy with no multi-step process; there are no validation checkpoints or feedback loops, which is appropriate but caps the workflow dimension below 3.

2 / 3

Progressive Disclosure

Well-organized into short labeled sections with no bundle files and no nested references, fitting the simple-skill allowance of scoring 3 with clean organization alone.

3 / 3

Total

11

/

12

Passed

Description

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A concise, third-person description that states concrete actions and gives explicit, natural 'Use when...' triggers tied to a distinctive untrusted-group-chat context. It clearly satisfies all four dimensions without padding.

DimensionReasoningScore

Specificity

Lists multiple concrete actions: 'Lists permitted and prohibited actions', 'blocks disallowed content types', and 'responds to permission queries', rather than vague language.

3 / 3

Completeness

Explicitly states what it does in the first sentence and provides an explicit 'Use when...' clause with multiple concrete trigger conditions.

3 / 3

Trigger Term Quality

Covers natural user phrasing such as 'joining a new group', 'unsure about rules, permissions, or boundaries', and 'public channel or untrusted group chat environment'.

3 / 3

Distinctiveness Conflict Risk

The shared/public group and untrusted-group-chat framing carves a distinct niche unlikely to trigger for unrelated skills.

3 / 3

Total

12

/

12

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Reviewed

Table of Contents