CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

convex-agent

Add an AI agent / RAG backend (@convex-dev/agent) to the Convex app.

Quality

65%

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

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/convex-agent/SKILL.md

The canonical home for this skill is convex-agent in openclaw/clawhub

SKILL.md
Quality
Evals
Security

Quality

Content

61%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 genuinely lean, well-structured overview that correctly prioritizes rules (gateway-first, keys in Convex env, never client-side). Its main weaknesses are the complete absence of executable code or commands and missing validation checkpoints in the workflow, which leave Claude to fill in the actual API calls from prior knowledge or docs.

Suggestions

Add a minimal copy-paste code snippet for the core path: defining the agent with `languageModel: convexGateway("provider/model")` in convex.config.ts and creating/streaming a thread from an action.

Insert validation checkpoints into the workflow (e.g., verify the agent component is registered after `npx convex dev`, and test one streamed message before adding RAG).

Move the RAG/embedding setup and thread API details into one or two short reference files (e.g. references/rag.md) instead of compressing them into single-line workflow steps.

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence with no padding or explanation of known concepts ("Install @convex-dev/agent + @convex-dev/ai-sdk-provider; add the agent component to convex.config.ts"). A minor deduction applies for the inline time-sensitive version pin "(needs convex 1.45+ on a Convex Cloud deployment, paid plan)" which is not placed in a deprecated/old-patterns section; otherwise this would be a 5.

4 / 5

Actionability

It names exact packages, the config file, and the `convexGateway("provider/model")` call, but contains zero executable code or commands — no install command, no agent-definition snippet, no thread-creation or embedding example. This matches anchor 3 ('Some concrete guidance but incomplete... missing key details') rather than 4's 'concrete code or commands with minor gaps'.

3 / 5

Workflow Clarity

The Workflow section lists a clear 5-step sequence, but there are no validation checkpoints (e.g., confirming the component registered or testing a streamed message), and steps like "Create threads + stream messages; persist history in Convex" compress multiple actions. This matches anchor 3 ('Steps listed but validation gaps; checkpoints missing or implicit'). The destructive/batch cap does not apply since this is an additive integration task.

3 / 5

Progressive Disclosure

The body is under 50 lines with well-organized Workflow and Rules sections, and no bundle files exist, so per the simple-skill note it could approach 5. However, the topic (agent API, RAG vector-index setup, env micro power) is complex enough that one-line compressed steps would benefit from one-level-deep reference files, matching anchor 4 ('Good structure... minor organization gaps').

4 / 5

Total

14

/

20

Passed

Description

53%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 is concise, third-person, and clearly anchored to a specific platform and package, giving it strong distinctiveness. However, it lists only a single action, omits capability synonyms users would naturally say, and entirely lacks a 'Use when...' trigger clause, capping completeness and trigger-term quality at the midpoint.

Suggestions

Add an explicit trigger clause, e.g. "Use when the user wants an in-app AI agent, chatbot, assistant, or RAG/vector search in a Convex app."

List 2-3 concrete capabilities in the description itself (durable threads, tool-calls, vector search/RAG) instead of the single phrase 'backend'.

Include natural synonyms such as chatbot, LLM chat, embeddings, and vector search so users' phrasing matches without knowing the package name.

DimensionReasoningScore

Specificity

The description names the domain and a concrete package ("Add an AI agent / RAG backend (@convex-dev/agent) to the Convex app") but lists only one action. It matches anchor 3 ('Names domain and 1-2 concrete actions, but not comprehensive'); it is not 4 because capabilities like durable threads, tool calls, or vector search are absent from the description.

3 / 5

Completeness

It has a clear 'what' (add an AI agent / RAG backend via @convex-dev/agent to a Convex app) but no 'Use when...' clause or equivalent explicit trigger guidance, which per the judging guidelines caps completeness at 3. It is not 2 because the 'what' is specific and unambiguous.

3 / 5

Trigger Term Quality

Includes natural terms users would say ("AI agent", "RAG", "Convex") but misses common variations and synonyms such as chatbot, assistant, LLM chat, vector search, or embeddings. This matches anchor 3 ('Some relevant keywords but missing common variations or synonyms') rather than 4's 'good keyword coverage'.

3 / 5

Distinctiveness Conflict Risk

The phrase "to the Convex app" plus the @convex-dev/agent package name creates a clear niche with low conflict risk, matching anchor 4 ('Mostly distinct; minor overlap risk with closely related skills'). It is not 5 because 'AI agent / RAG backend' without trigger phrasing has minor overlap with generic agent-building skills.

4 / 5

Total

13

/

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

FacundoLucci/autonomous-buyer

Path

.agents/skills/convex-agent/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.