safety-guard

Use this skill to prevent destructive operations when working on production systems or running agents autonomously.

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No known issues

Fix and improve this skill with Tessl

tessl review fix ./skills/safety-guard/SKILL.md

Quality

Content

72%

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

The content is concise and well-structured with concrete commands and a useful pattern list, but actionability and workflow clarity are held back by a descriptive rather than executable implementation section and the absence of a recovery feedback loop.

Suggestions

Add the actual PreToolUse hook JSON configuration so the Implementation section is copy-paste ready instead of a description of the mechanism.

Document the error-recovery flow when a user denies a destructive command (what state persists, how to safely re-run) to provide an explicit feedback loop.

Frame the three modes as a decision workflow with explicit selection criteria (when to choose Careful vs Freeze vs Guard) rather than a flat menu.

Dimension	Reasoning	Score
Conciseness	The body is lean and well-organized, with concrete watched-pattern lists and brief mode descriptions; it does not explain concepts Claude already knows, and every section earns its place, matching the 'lean and efficient; every token earns its place' anchor.	3 / 3
Actionability	It provides concrete user-facing slash commands ('/safety-guard freeze src/components/', '/safety-guard guard --dir src/api/ --allow-read-all', '/safety-guard off') and a concrete pattern list, but the Implementation section only describes the PreToolUse hook mechanism ('Uses PreToolUse hooks to intercept...') without providing the actual hook configuration, leaving key setup details missing; this fits 'some concrete guidance but incomplete'.	2 / 3
Workflow Clarity	The Mode 1 interception flow (detect -> show -> ask for confirmation -> suggest alternative) has an explicit confirmation checkpoint, but the three modes are presented as a menu of alternatives rather than a sequenced workflow, and there is no error-recovery feedback loop, matching 'steps listed but validation gaps; checkpoints implicit'.	2 / 3
Progressive Disclosure	It is a self-contained, well-organized file with clear sections (When to Use, How It Works with three modes, Implementation, Integration) and no need for external references; per the simple-skills note, well-organized sections with no required references score 3.	3 / 3
	Total	10 / 12 Passed

Description

75%

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

The description clearly conveys both purpose and trigger conditions and occupies a distinct niche, but the action language is abstract ('prevent') and the trigger-term coverage could include more of the phrasings users actually say.

Dimension	Reasoning	Score
Specificity	It names the domain ('destructive operations', 'production systems', 'running agents autonomously') and a single action ('prevent'), but does not list multiple concrete actions, matching the 'names domain and some actions, but not comprehensive' anchor rather than the multi-action level 3.	2 / 3
Completeness	It explicitly states what ('prevent destructive operations') and when ('when working on production systems or running agents autonomously'), with an explicit trigger clause, matching the 'clearly answers both what AND when with explicit triggers' anchor; it is not the level below because the when-clause is explicit rather than implied.	3 / 3
Trigger Term Quality	'production systems' and 'running agents autonomously' are reasonably natural trigger terms, but coverage misses common variations a user might actually say ('rm -rf', 'safe mode', 'don't delete'), fitting 'some relevant keywords but missing common variations'.	2 / 3
Distinctiveness Conflict Risk	It targets a distinct niche (destructive-operation prevention on production and autonomous-agent systems) with triggers unlikely to fire for unrelated skills, matching the 'clear niche with distinct triggers' anchor.	3 / 3
	Total	10 / 12 Passed

Validation

93%

Warnings & errors only

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

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: affaan-m/ECC
Commit: 754b8dd

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.