CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

accidental-data-loss-prevention

**STOP AND VERIFY**: Before running any command or tool that results in irreversible data loss, you MUST obtain explicit user consent. When in doubt, ask. It is better to wait for confirmation than to accidentally delete production data or critical project assets. Use this for: - SQL: DROP TABLE/VIEW/SCHEMA/DATABASE, TRUNCATE, or broad DELETE (missing WHERE or using 1=1). - Cloud Storage: gsutil rm or gcloud storage rm targeting production data or critical buckets. - Infrastructure: gcloud projects delete, deleting Spanner/BigQuery/Dataproc resources, deleting secrets, or KMS key destruction.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

78%Weight 40%Scale 1-5

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

The body is a lean, well-structured consent procedure whose validation gate (explicit affirmative consent) is the explicit core of the workflow, well matched to its single purpose. Minor improvements would be a consent-request template and guidance for denial or ambiguous responses.

DimensionReasoningScore

Conciseness

The body is lean with no explanations of concepts Claude already knows, but the CAUTION callout ('**STOP AND VERIFY**: Before running any command or tool that results in irreversible data loss, you **MUST** obtain explicit user consent') duplicates the frontmatter description verbatim — a minor instance of redundancy that could be trimmed, matching anchor 4 rather than 5's 'every token earns its place'.

4 / 5

Actionability

Concrete, executable instruction-style guidance: 'Do **not** execute the command', explain 'the **impact**' and '**why**', request 'their **explicit approval**', and 'Only proceed if the user provides clear, affirmative consent' — but no sample consent-request template and no handling of denial or ambiguous responses, matching anchor 4 ('mostly executable guidance... minor gaps') rather than 5.

4 / 5

Workflow Clarity

A clear 3-step sequence ('Halt Execution' → 'Request Consent' → 'Wait') with an explicit validation checkpoint (clear affirmative consent required before proceeding), so the destructive-operation cap for missing validation does not apply. Not 5 because there is no feedback loop for error recovery — the user-denial or ambiguous-consent path is unspecified.

4 / 5

Progressive Disclosure

The body is under 50 lines, self-contained with no need for external references (no references/, scripts/, or assets/ files exist), and organized into a CAUTION callout plus a well-structured 'Mandatory Procedure' section — per the scoring notes, a simple skill in this form scores 5.

5 / 5

Total

17

/

20

Passed

Description

78%Weight 40%Scale 1-5

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 communicates a consent gate for irreversible data-loss commands, with an explicit 'Use this for' clause covering concrete SQL and GCP operations. Its main weaknesses are the second-person opening (which costs a specificity point) and missing natural synonyms/variations beyond GCP-specific tooling.

Suggestions

Rewrite the opening in third person (e.g., 'Halts execution and requires explicit user consent before any command causing irreversible data loss') to avoid the second-person specificity penalty.

Add natural trigger synonyms and variations users might say, such as 'delete the table', 'empty the bucket', or 'remove production data', and consider non-GCP equivalents (aws s3 rm, kubectl delete) for broader keyword coverage.

DimensionReasoningScore

Specificity

Lists several concrete operations ('DROP TABLE/VIEW/SCHEMA/DATABASE, TRUNCATE, or broad DELETE (missing WHERE or using 1=1)', 'gsutil rm or gcloud storage rm', 'gcloud projects delete... KMS key destruction') with only minor coverage gaps (no filesystem rm, AWS/Azure equivalents) — a base 4 — but the opening 'you MUST obtain explicit user consent' is second-person voice, which the guidelines penalize by reducing specificity by 1.

3 / 5

Completeness

The 'what' is explicit ('you MUST obtain explicit user consent' before irreversible data loss) and the 'when' is explicit via 'Use this for:' with concrete command-level trigger cases across SQL, Cloud Storage, and Infrastructure — clearly matches anchor 5; not 4 because the when-clause is already fully explicit and specific, so the missing-'Use when' cap does not apply.

5 / 5

Trigger Term Quality

'TRUNCATE', 'DROP TABLE', 'gsutil rm', and 'gcloud projects delete' are phrases users actually type, giving good keyword coverage, but common natural variations and synonyms (plain 'delete the table', 'empty the bucket', non-GCP cloud equivalents) are missing — matches anchor 4 ('good keyword coverage; a few natural terms missing'), not 5's comprehensive synonym coverage.

4 / 5

Distinctiveness Conflict Risk

Clear niche (consent gate for irreversible data-loss operations on SQL/GCP) with distinct command-level triggers, but 'When in doubt, ask' widens the trigger surface and closely related safety/confirmation skills could overlap — matches anchor 4 ('mostly distinct; minor overlap risk with closely related skills') rather than 5's minimal conflict risk.

4 / 5

Total

16

/

20

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.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

saski/arnesto

Path

.agents/skills/accidental-data-loss-prevention/SKILL.md

Reviewed

Table of Contents

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.