CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

a2a-protocol

How agent-native apps discover, authenticate with, and call each other over A2A. Use when connecting one app to another, debugging a remote agent that won't authenticate, exposing skills to peers, or calling agents from scripts.

Quality

82%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

78%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.

Highly actionable and information-dense reference content: executable code for every calling style, concrete auth semantics, and explicit validation via the peer probe. Its main weakness is monolithic structure — with no references directory, ~530 lines of mixed policy, provider integration, and API detail are inlined where progressive disclosure would split them into focused reference files.

Suggestions

Move the hosted-provider integration details (Foundry, Gemini Enterprise, Anthropic Managed Agents) into a references/hosted-providers.md file, leaving a one-line summary and pointer per provider in SKILL.md.

Move the full type-export list and custom-mount/A2AConfig API detail into a references/api.md file, keeping only the most common callAgent/invokeAgentAction examples inline.

Consolidate the repeated 'fix the delegation path rather than bypassing A2A' guidance into the Rule section and reference it from the Calling section to trim duplicated tokens.

DimensionReasoningScore

Conciseness

Nearly every sentence carries project-specific operational facts Claude cannot know (secrets, endpoints, error codes, policy rules), with no padding of general concepts. Minor trimming is possible — e.g., the 'fix the delegation path, don't bypass it' guidance is restated in both the Rule and Calling sections — which keeps it below anchor 5.

4 / 5

Actionability

Copy-paste-ready code for callAgent(), invokeAgentAction(), A2AClient (discover/send/stream), approvedActions, and the custom mount plugin, plus complete JSON manifests, exact endpoints, and failure-mode tables. The examples cover the common calling patterns end-to-end, matching the anchor-5 example.

5 / 5

Workflow Clarity

Sections follow a coherent sequence (eligibility rule → A2A already mounted → register peer → probe → auth → advertise → call) and include an explicit validation step: the probe endpoint returning 'reachable' and 'authorized' as independent fields, plus the 401/503 diagnosis table. Not 5 because the flow is presented as topical sections rather than an explicitly ordered procedure with error-recovery loops for the registration-to-first-call path.

4 / 5

Progressive Disclosure

The file has clear section headers and tables, but there are no bundle files at all (no references/, scripts/, or assets/ exist) and everything is inlined in one ~530-line SKILL.md — hosted-provider integration details, the full type-export listing, and the custom-mount API all belong in separate one-level-deep reference files. This matches anchor 3: structure present, but content that should be separate is inline.

3 / 5

Total

16

/

20

Passed

Description

83%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: third-person, concrete about capabilities, and anchored by an explicit 'Use when' clause with natural trigger phrases covering connect, debug, expose, and script scenarios. The only weakness is slightly thin synonym coverage (e.g., 'agent-to-agent', 'peer') in the trigger terms.

DimensionReasoningScore

Specificity

Names concrete actions — 'discover, authenticate with, and call each other over A2A' plus 'exposing skills to peers' — giving several specific capabilities with only minor coverage gaps (no mention of task lifecycle or streaming). It sits above anchor 3 (only 1-2 actions) but below anchor 5's comprehensiveness.

4 / 5

Completeness

Explicitly answers both questions: 'what' is the opening clause (discover, authenticate, call over A2A) and 'when' is a concrete 'Use when…' clause with four specific trigger scenarios, matching the anchor-5 pattern. Not 4, because the 'when' guidance is already explicit and specific rather than improvable.

5 / 5

Trigger Term Quality

The 'Use when' clause supplies natural phrases users would actually say: 'connecting one app to another', 'debugging a remote agent that won't authenticate', 'exposing skills to peers', 'calling agents from scripts'. Coverage is good but misses common synonyms like 'agent-to-agent' or plain 'peer', keeping it below anchor 5.

4 / 5

Distinctiveness Conflict Risk

A clear niche — inter-app A2A protocol, remote-agent registration and authentication — with distinct triggers like 'remote agent that won't authenticate'. Minor overlap risk remains with generic agent-delegation or API-connection skills, so it does not reach anchor 5's minimal conflict risk.

4 / 5

Total

17

/

20

Passed

Validation

75%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 12 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (542 lines); consider splitting into references/ and linking

Warning

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

12

/

16

Passed

Repository

BuilderIO/agent-native

Path

.agents/skills/a2a-protocol/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.