Name: ax-agent-observability
Rating: 68.8 (1 reviews)
Author: ax-llm

ax-agent-observability

This skill helps an LLM generate correct AxAgent observability code using @ax-llm/ax. Use when the user asks about axGlobals.onUsage, usageContext, centralized or multi-tenant usage accounting, actorTurnCallback, onContextEvent, agentStatusCallback, onFunctionCall, reportSuccess, reportFailure, getChatLog(), getUsage(), resetUsage(), debug traces, progress updates, or telemetry for AxAgent runs.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

77%

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

A thorough, highly actionable reference with executable code, type signatures, and a clear hook-selection workflow plus a guardrail checklist. Its main weaknesses are a duplicated type block and a monolithic structure that could split per-hook detail into reference files.

Suggestions

Remove the duplicated actorTurnCallback type block (or collapse the 'deprecated alias' into a one-line note) to reclaim tokens.

Split the per-hook type references and full event-shape detail into files under references/ (e.g. references/usage-observer.md, references/callbacks.md) and point to them from SKILL.md so the main file stays an overview.

De-duplicate points that appear in both the 'What it gives you' and 'Important' lists for actorTurnCallback.

Dimension	Reasoning	Score
Conciseness	The body is dense and mostly lean, but the actorTurnCallback type signature is duplicated verbatim (a second block labeled 'deprecated alias') and the 'What it gives you' and 'Important' lists restate some points, so it could be tightened.	2 / 3
Actionability	Every hook ships with complete, copy-paste-ready TypeScript examples and full type signatures (e.g. the onFunctionCall call shape, getUsage()/getStagedUsage() snippets), giving fully executable guidance.	3 / 3
Workflow Clarity	The 'Choose The Smallest Hook' section gives a clear, sequenced selection procedure and the 'Do Not Generate' section is an explicit guardrail checklist; no destructive/batch operation is involved so missing validation checkpoints do not cap the score.	3 / 3
Progressive Disclosure	Sections are well-organized, but the ~390-line body is monolithic with all type references inline and no bundle files to offload detail; deeper material is only lightly signaled via cross-skill pointers like 'also read ax-ai'.	2 / 3
	Total	10 / 12 Passed

Description

90%

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 with an explicit Use-when clause and rich, natural trigger terms; it uses third-person voice and avoids fluff. The only weakness is that the capability is expressed as a single generate-code action rather than enumerated concrete actions.

Suggestions

Replace or augment 'generate correct AxAgent observability code' with a few concrete verbs (e.g. 'wire up usage observers, register actor-turn/context/status callbacks, and read chat logs and usage') so multiple specific actions are visible.

Consider trimming a few of the most obscure API tokens from the trigger list to keep the description focused on terms users actually say.

Dimension	Reasoning	Score
Specificity	It names the domain ('AxAgent observability code using @ax-llm/ax') and one concrete action ('generate correct ... code'), but the skill's action surface is framed as a single verb rather than multiple distinct concrete actions.	2 / 3
Completeness	It states what it does ('helps an LLM generate correct AxAgent observability code') and an explicit 'Use when the user asks about ...' clause covering both what and when.	3 / 3
Trigger Term Quality	A long, natural list of terms a developer would actually say — 'axGlobals.onUsage', 'actorTurnCallback', 'onContextEvent', 'getChatLog()', 'debug traces', 'telemetry' — gives strong coverage of likely triggers.	3 / 3
Distinctiveness Conflict Risk	Scoped tightly to '@ax-llm/ax' AxAgent observability with specific API names and 'for AxAgent runs', making it unlikely to trigger for sibling skills like ax-agent or ax-agent-rlm.	3 / 3
	Total	11 / 12 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: ax-llm/ax
Path: website/static/typescript/.well-known/agent-skills/ax-agent-observability/SKILL.md
Commit: 7911e27

Reviewed: 5 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.