ax-rust-agent-observability

Use when writing Rust code with `axllm` for agent tracing, centralized and multi-tenant usage accounting, action logs, runtime diagnostics, replay, and production debugging.

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./website/static/rust/.well-known/agent-skills/ax-rust-agent-observability/SKILL.md

Quality

Content

65%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.

A lean, information-dense body with concrete API names, snippets, and sensible guardrails. Its main weaknesses are the unsequenced usage-observer guidance and the inline API-surface dump that should live in (or defer to) a reference file.

Suggestions

Sequence the Centralized Usage Observer section as explicit numbered steps — attach `usageContext` defaults, register the observer, forward, drain the queue out of band, clear via `set_usage_observer(None)` on teardown — so the setup workflow is unambiguous.

Replace the flat 'Relevant API Surface' identifier dump with a short pointer to the package's `API.md` / `axir-api.json`, keeping only the handful of names the recipes in the body actually use.

Make the two code snippets compilable (imports and defined `llm`/`inputs`/`usage_queue`), or explicitly label them as fragments and route readers to the runnable `usage_observer.rs` example for the full program.

Dimension	Reasoning	Score
Conciseness	The body is efficient — terse bullet facts and focused snippets with no explanations of concepts Claude already knows — but the ~40-identifier flat dump in 'Relevant API Surface' partially duplicates the cited `API.md`/`axir-api.json` and could be trimmed, matching the 'minor instances that could be trimmed' level-4 anchor rather than the every-token-earns-its-place level-5 anchor.	4 / 5
Actionability	Concrete guidance dominates: the core pattern and `set_usage_observer` snippets, specific rules ('Attach `usageContext` in call or agent-forward option maps', 'attributes are shallow-merged'), and a pointer to a runnable example (`src/examples/rust/generation/usage_observer.rs`) — but the snippets are not copy-paste compilable (undefined `llm`, `usage_queue`, no imports), so it sits at 'mostly executable with minor gaps' rather than fully executable.	4 / 5
Workflow Clarity	The pieces of the main workflow (attach usage context, register observer, enqueue synchronously, clear on teardown) all exist but are scattered across unsequenced bullets with no ordered steps or explicit checkpoints for the usage-accounting setup; guardrails like 'Start from package examples for exact native syntax' act only as implicit checkpoints, matching the level-3 anchor.	3 / 5
Progressive Disclosure	Sections are well organized and the skill is self-contained (no bundle files exist to reference), but the 'Relevant API Surface' section inlines a bulk identifier listing that belongs in a separate reference file — especially since `API.md` and `axir-api.json` are already named as package docs — fitting the 'content that should be separate is inline' level-3 anchor.	3 / 5
	Total	14 / 20 Passed

Description

78%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, tight description: explicit 'Use when' trigger, specific niche with the package name, and a concrete capability list. Main gaps are the merged what/when phrasing and missing common synonyms like 'observability' and 'telemetry'.

Suggestions

Split the description into a what clause and a when clause, e.g. 'Register process-wide usage observers, attribute model calls by tenant/user/request, and inspect traces and action logs in Rust with `axllm`. Use when the user mentions agent observability, usage accounting, tracing, or production debugging of `axllm` agents.'

Add natural synonyms users would say — 'observability', 'telemetry', 'token usage', 'monitoring' — to widen trigger coverage.

Dimension	Reasoning	Score
Specificity	The description lists several concrete capability areas — 'agent tracing, centralized and multi-tenant usage accounting, action logs, runtime diagnostics, replay, and production debugging' — but they are noun phrases rather than the concrete actions the skill performs (e.g., registering a usage observer, attaching usage context), which keeps it below the comprehensive level-5 anchor and clearly above the 1-2-action level-3 anchor.	4 / 5
Completeness	Both what and when are explicitly present via 'Use when writing Rust code with `axllm` for agent tracing, ...' — an explicit trigger clause plus a capability list — but what and when are merged into a single clause rather than separate statements with concrete trigger phrases, so it fits the level-4 anchor better than the level-5 exemplar.	4 / 5
Trigger Term Quality	Natural terms users would say are present ('Rust', 'agent tracing', 'usage accounting', 'action logs', 'replay', 'production debugging', 'axllm'), giving good keyword coverage, but common variations like 'observability', 'telemetry', 'monitoring', and 'token usage' are missing — more coverage than the level-3 anchor, not the synonym-complete level-5 anchor.	4 / 5
Distinctiveness Conflict Risk	The trigger is pinned to a specific language (Rust), a specific package name (`axllm`), and a specific niche (agent observability/usage accounting), giving it a clear niche with distinct triggers and minimal conflict risk with generic skills.	5 / 5
	Total	17 / 20 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/rust/.well-known/agent-skills/ax-rust-agent-observability/SKILL.md
Commit: 3f630a0

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.