CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

data

Validate PsychSift imports, seeds, transformations, reconciliation, deduplication, integrity, provenance, and rollback controls. Use for dataset changes, bulk operations, registry content, or data migration logic.

Quality

86%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Medium

Suggest reviewing before use

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.

An extremely lean, well-sequenced checklist for a data-validation workflow with genuine checkpoints and an approval gate for risky operations. The gap is actionability: it tells Claude what to verify but not how — no commands, fixture layouts, or dry-run mechanics.

Suggestions

Add one short example of a concrete test invocation, e.g. a sample command or fixture setup showing how to run a local dry run with malformed/duplicate inputs.

Specify what 'Measure counts and invariants' looks like in practice — which counts and which invariants to compare before/after (e.g. row counts per table, uniqueness of identifiers).

Include a brief feedback-loop instruction for step 4: what to do when invariants fail or records are rejected (fix source, rerun, re-measure) to close the workflow's error-recovery gap.

DimensionReasoningScore

Conciseness

Six lean numbered lines with zero padding and no explanation of concepts Claude already knows; every token earns its place. This is a model of token efficiency for a checklist-style skill.

5 / 5

Actionability

Guidance is specific about what to check ("malformed, missing, duplicate, conflicting, oversized, partial, and rerun inputs", "counts and invariants") but gives no executable specifics — no commands, fixture examples, or how to run the local dry runs — leaving key execution details missing as in anchor 3. As an instruction-only skill absence of code is not penalized per se, but the how-to is still underspecified.

3 / 5

Workflow Clarity

Clear sequence (identify → test → check properties → measure → approval gate → report) with explicit validation checkpoints ("Measure counts and invariants before and after local dry runs") and a safety gate for live operations. Falls short of anchor 5 because there is no feedback loop — nothing says what to do when invariants fail or validation rejects records.

4 / 5

Progressive Disclosure

A sub-50-line skill with no need for external references and no bundle files present; the single header plus well-organized numbered list satisfies the simple-skill exception for a top score.

5 / 5

Total

17

/

20

Passed

Description

87%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.

A strong description: concrete coverage of the domain's operations, an explicit 'Use for' trigger clause with natural phrases, and a distinct niche anchored by the PsychSift name. The only softness is that the capability list leans on one verb governing a noun enumeration rather than varied concrete actions.

DimensionReasoningScore

Specificity

Enumerates many specific objects of validation — "imports, seeds, transformations, reconciliation, deduplication, integrity, provenance, and rollback controls" — but the action vocabulary is a single verb ("Validate") over a noun list rather than multiple distinct concrete actions like the score-5 anchor. Minor gaps only, so above the 1-2-action level of anchor 3.

4 / 5

Completeness

Explicitly answers both: a concrete what ("Validate PsychSift imports, seeds, transformations, reconciliation, deduplication, integrity, provenance, and rollback controls") and an explicit when with concrete triggers ("Use for dataset changes, bulk operations, registry content, or data migration logic"), matching the structure of the score-5 anchor.

5 / 5

Trigger Term Quality

"dataset changes, bulk operations, registry content, or data migration logic" are natural phrases a user would say when needing this skill, alongside domain terms like "imports" and "seeds". Missing common synonyms and extensions, so it falls just short of the comprehensive anchor 5.

4 / 5

Distinctiveness Conflict Risk

"PsychSift" names a specific system, giving a clear niche with distinct trigger terms; it is unlikely to fire for unrelated skills. Not score 4, since the named system plus the enumerated operation types leave minimal overlap risk.

5 / 5

Total

18

/

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

BigSimmo/PsychSift

Path

.agents/skills/data/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.