convex-improve-convex-plugin

Send this coding session's transcript to the Convex team for an AI post-mortem that improves the quickstart system.

Quality

77%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/convex-improve-convex-plugin/SKILL.md

Quality

Content

85%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 content is a tight, actionable runbook: a concrete command, explicit error-recovery for the consent gate, and clear output markers, organized into clean Workflow and Rules sections. Its only deductions are minor — repeated consent-gate text and an undefined <anteater> placeholder that keeps the curl command from being fully copy-paste ready.

Suggestions

State the consent rule once (in Rules) and have Workflow step 2 reference it, removing the triple repetition between the intro, step 2, and the first rule.

Define how "<anteater>" resolves (env var, config value, or literal URL) so the curl command is copy-paste executable as written.

Dimension	Reasoning	Score
Conciseness	The ~20-line body is lean and free of concept over-explanation, but the consent gate is stated three times (intro paragraph, Workflow step 2, and Rules bullet 1), so some tokens could be trimmed. Anchor 4 ('minor instances that could be trimmed'), not 5's 'every token earns its place'.	4 / 5
Actionability	The body gives a concrete curl command, a re-run command with --consent, exit-code handling (CONSENT_REQUIRED, exit 4), and specific output markers to watch. Not 5 because "<anteater>" is an unresolved placeholder — the actual endpoint URL is never defined, so the command is not copy-paste ready.	4 / 5
Workflow Clarity	Four numbered steps in clear sequence with explicit validation checkpoints (output markers REVIEW_SOURCE / REVIEW_SUBMITTED / REVIEW_DONE) and a concrete error-recovery loop (exit 4 → ask user → re-run with --consent; REVIEW_NO_TRANSCRIPT handling). This matches the anchor requiring explicit validation steps plus feedback loops for error recovery.	5 / 5
Progressive Disclosure	The skill is under 50 lines with no bundle files (no references/, scripts/, or assets/ exist), and the body is organized into a brief intro, Workflow, and Rules sections with nothing that belongs in a separate file. Per the simple-skill guideline, this scores 5.	5 / 5
	Total	18 / 20 Passed

Description

57%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 states a specific, distinctive capability in a niche no other skill occupies, but it entirely lacks a 'use when' trigger clause and natural trigger phrases, and it covers only a single action of the skill's actual capability set. It reads as a clear 'what' with a missing 'when'.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user wants to share this coding session with the Convex team for quickstart feedback, or asks for a Convex post-mortem or quickstart review."

Include natural user phrasings and synonyms — "share session transcript", "improve the quickstart", "Convex review" — so the trigger terms match what users actually say.

Mention the follow-up capability (summarizing severity-ranked findings) so the description covers the skill's actions rather than only the initial send step.

Dimension	Reasoning	Score
Specificity	"Send this coding session's transcript to the Convex team" names one concrete action plus its purpose ("AI post-mortem that improves the quickstart system"), matching the 1-2-concrete-actions anchor. It is not 4 because it omits the consent flow and findings-summarization capabilities the skill actually has.	3 / 5
Completeness	The 'what' is clear (sends the session transcript to the Convex team for an AI post-mortem), but there is no 'when' clause or equivalent trigger guidance at all, which per the judging guidelines caps completeness at 3. Not 2 because the 'what' half is specific and accurate.	3 / 5
Trigger Term Quality	Domain keywords like "transcript", "Convex", "post-mortem", and "quickstart" are present, but natural user phrasings and synonyms (e.g. "share my session", "review the quickstart") are missing. It sits at 'some relevant keywords but missing common variations', not the good-coverage anchor.	3 / 5
Distinctiveness Conflict Risk	"Send this coding session's transcript to the Convex team for an AI post-mortem that improves the quickstart system" is a clear niche with distinct triggers; virtually no other skill would match this description. Not 4 because there is no realistic overlap with related skills.	5 / 5
	Total	14 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: openclaw/clawhub
Path: .agents/skills/convex-improve-convex-plugin/SKILL.md
Commit: 0180b56

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.