ax-cpp-agent-observability

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

Quality

88%

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

Quality

Content

82%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 an efficient, well-guarded reference for a generated C++ package: novel facts only, executable core snippets, exact symbol names, and pointers to runnable examples. Its main weaknesses are minor: a few prerequisite code paths (service/client construction, usageContext attachment) are prose-only, and the large inline API listing keeps organization just below the top anchor.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes competence: dense factual bullets ('Real network support: yes', observer semantics, merge/override rules) with zero explanation of concepts Claude already knows. The long API symbol list is an inventory for lookup, not padded prose, so every line earns its place.	5 / 5
Actionability	The Core Pattern ('axllm::agent(...) / helper.forward(llm, {...})') and the set_usage_observer snippet are concrete and near copy-paste ready, with exact symbols and a runnable example path ('src/examples/cpp/generation/usage_observer.cpp'). Not a 5: minor gaps remain — constructing the `llm` service (including a client) and attaching `usageContext` are described in prose but never shown in code.	4 / 5
Workflow Clarity	The Guardrails section supplies a usable sequence: start from package examples for exact syntax, use no-key examples for deterministic local checks (a validation mechanism), provider-api examples only with credentials, and treat AxIR as source of truth when docs conflict. Not a 5: this direction is implicit rather than an explicit ordered workflow, and no validation step exists for the usage-observer integration path.	4 / 5
Progressive Disclosure	Sections are clear and package resources (API.md, axir-api.json, axir-capabilities.json, examples/) are signaled one level deep under 'Package Facts'; no bundle files exist to nest. Not a 5: the ~50-symbol 'Relevant API Surface' dump is inlined monolithically rather than split or summarized, a minor organization gap against the anchor-5 'content appropriately split' bar.	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 description: explicit 'Use when' trigger, third-person voice, concrete capability list, and a clearly delineated niche (C++ + axllm). The only gap is missing common synonyms such as 'observability', 'telemetry', and 'monitoring' that users would naturally say, which keeps trigger term quality just short of comprehensive.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete capability areas — 'agent tracing, centralized and multi-tenant usage accounting, action logs, runtime diagnostics, replay, and production debugging' — with comprehensive coverage matching the skill's scope. Not a 4: the action list is comprehensive, not just 'several specific actions' with gaps.	5 / 5
Completeness	An explicit 'Use when writing C++ code with `axllm`' trigger clause answers 'when', and the enumerated purposes ('agent tracing ... production debugging') explicitly answer 'what'. Both are concrete and explicit, matching the anchor-5 pattern rather than the anchor-4 case where 'when' could be more specific.	5 / 5
Trigger Term Quality	Natural phrases like 'agent tracing', 'usage accounting', 'runtime diagnostics', 'replay', and 'production debugging' give good keyword coverage anchored to 'C++' and the package name 'axllm'. Not a 5: common synonyms a user might say — 'observability', 'telemetry', 'monitoring' (notably absent even though the skill title contains 'Observability') — are missing.	4 / 5
Distinctiveness Conflict Risk	The description is pinned to a narrow niche — C++ code with the specific `axllm` package for observability tasks — giving it distinct triggers with minimal overlap risk against generic C++ or other-language skills. No neighboring anchor fits better.	5 / 5
	Total	19 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 1 suspicious	Warning

	Total	14 / 16 Passed

Repository: ax-llm/ax
Path: website/static/cpp/.well-known/agent-skills/ax-cpp-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.