CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

create-and-cleanup-migration

Use for end-to-end ClawHub Convex production migrations, backfills, destructive cleanups, and one-off maintenance functions that must be created, validated, shipped, run, verified, then removed after completion.

Quality

89%

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

81%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 excellent instruction-only skill body: a seven-phase lifecycle with operator gates, dry-run-first safety rules, verification steps, and concrete validation commands. Weaknesses are minor — some deliberate repetition of the confirmation gates and delegation of the actual migration command shapes to companion skills rather than showing them.

Suggestions

Consolidate the repeated confirmation-gate language (stated in Safety Rules, Phase 5, and Phase 6) into a single authoritative statement to trim token usage without losing safety.

Include one concrete example command shape for a production dry run and apply (e.g., the `@convex-dev/migrations` runner invocation) so the core steps are executable without consulting companion skills.

Move the detailed per-phase checklists (e.g., Phase 5 result-presentation fields and Phase 7 cleanup lists) into a one-level-deep reference file to shorten the always-loaded SKILL.md.

DimensionReasoningScore

Conciseness

The body is dense, checklist-driven, and assumes Convex competence — no space is spent explaining concepts Claude already knows. However, the destructive-apply and cleanup-removal confirmation gates are each stated two or three times (Safety Rules, Phase 5 step 4, Phase 6 step 1), and bounded-batch resumption appears in both Phase 5 and Phase 6; this deliberate but trimmable duplication keeps it at anchor 4 ('minor instances... that could be trimmed') rather than 5.

4 / 5

Actionability

Concrete executable commands are given for validation (`bunx convex codegen`, `bunx tsc --noEmit`, `bun run ci:static`, `bunx convex dev --once`, `convex run`) plus specific result fields (`isDone`, `dryRun`, resume cursors, confirmation token), matching 'mostly executable guidance with minor gaps'. Not a 5 because the actual dry-run/apply command shapes are delegated to `@convex-dev/migrations` and companion skills rather than shown, so nothing is copy-paste ready for the core production steps.

4 / 5

Workflow Clarity

Seven clearly sequenced phases with explicit validation checkpoints at every gate: local validation ladder before PR, deploy success before dry run, dry-run results presented before apply, explicit user confirmation before both apply and cleanup removal, verification proving zero remaining targets, and resume-until-isDone feedback loops with documented safety caps. This matches anchor 5 (explicit validation steps, error-recovery loops, checklists for a complex destructive process); the destructive-operation cap of 3 does not apply since validation is pervasive.

5 / 5

Progressive Disclosure

The body is well organized into scannable phase sections and appropriately delegates detail to companion skills (`convex-migration-helper`, `clawhub-convex`, `clawhub-production-release`) rather than inlining it. It stays at anchor 4 rather than 5 because at ~170 lines it is a single monolithic file with no one-level-deep bundled references (no references/ or scripts/ exist), and some per-phase detail checklists could be split out to shorten the always-loaded overview.

4 / 5

Total

17

/

20

Passed

Description

92%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, well-scoped description in imperative 'Use for...' form that names the domain, multiple concrete operation types, the full lifecycle verbs, and explicit trigger scenarios. The only minor gap is a few natural synonyms (schema narrowing, table reshaping) that the body covers but the description omits.

Suggestions

Consider adding one or two natural synonyms from the body — e.g., 'schema narrowing' or 'table reshaping' — to broaden the trigger-term coverage.

Optionally name the companion entry point (e.g., 'starts with convex-migration-helper') in the description to further sharpen distinctiveness against neighboring migration skills.

DimensionReasoningScore

Specificity

The description enumerates multiple concrete job types ("end-to-end ClawHub Convex production migrations, backfills, destructive cleanups, and one-off maintenance functions") and the full action lifecycle ("created, validated, shipped, run, verified, then removed after completion"), giving comprehensive coverage of what the skill handles. Anchor 4 falls short because there are no real gaps in coverage; anchor 4's 'minor gaps' don't apply to this lifecycle-complete enumeration.

5 / 5

Completeness

'Use for end-to-end ClawHub Convex production migrations, backfills...' explicitly answers 'when' with concrete trigger phrases, while 'created, validated, shipped, run, verified, then removed after completion' explicitly answers 'what' the skill drives. This matches anchor 5's requirement for both a clear what and an explicit when with concrete triggers; anchor 4's weaker 'when could be more explicit' does not apply.

5 / 5

Trigger Term Quality

Natural domain terms a user would say — "production migrations", "backfills", "destructive cleanups", "one-off maintenance functions" — are present, matching 'good keyword coverage; a few natural terms missing'. It is not a 5 because synonyms that appear in the body (e.g., 'schema narrowing', 'table reshaping', 'data migration') are absent from the description itself.

4 / 5

Distinctiveness Conflict Risk

The description carves out a clear niche — the end-to-end create-through-cleanup lifecycle for ClawHub Convex production changes — with distinct trigger terms and minimal overlap risk. Anchor 4's 'minor overlap risk with closely related skills' is not met: even adjacent migration-helper or release skills are distinguishable from this lifecycle-with-removal framing.

5 / 5

Total

19

/

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

openclaw/clawhub

Path

.agents/skills/create-and-cleanup-migration/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.