agentmail

Use your assigned AgentMail inbox to read email tasks, explicitly send or reply, and check delivery. Provided automatically by your inbox assignment.

Quality

61%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/agentmail/SKILL.md

Quality

Content

68%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 actionable and token-efficient with concrete CLI and HTTP guidance, but the multi-step send/verify workflow is conveyed in prose without an explicit numbered sequence or validation checkpoint. Content is appropriately self-contained with light sectioning.

Suggestions

Convert the discover→read→send/reply→inspect-delivery flow into a numbered list with an explicit 'Validate delivery' checkpoint (e.g. '4. Inspect publication outcome; only treat as sent once delivery confirms') to raise workflow clarity.

Remove the duplication between the prose endpoint/field descriptions and the HTTP API reference table — keep the table as the single source for endpoints and link fields once.

Add one complete request.json example for a new send and a reply so the send/reply guidance is fully copy-paste ready.

Dimension	Reasoning	Score
Conciseness	The body is dense and assumes competence without explaining what email or APIs are, but the HTTP API reference table and trailing paragraph repeat endpoints and required fields already stated in the prose, a minor trimmable redundancy keeping it just below lean anchor 5.	4 / 5
Actionability	Concrete copy-paste commands ('paperclipai email inboxes', '... thread "$PAPERCLIP_TASK_ID"', '... send --file <request.json>', '... delivery <publicationId>') and explicit HTTP method+path coverage make it mostly executable; the minor gap is no complete request.json payload example, so it stops short of anchor 5.	4 / 5
Workflow Clarity	The discover→read→send/reply→inspect-delivery→retry flow is present in prose with an implicit delivery checkpoint and idempotency guidance, but it is not a numbered sequence and the verification checkpoint is implicit rather than an explicit validate-then-proceed step, matching anchor 3.	3 / 5
Progressive Disclosure	No bundle files exist and the skill is self-contained; a single '## HTTP API reference' section with a table gives good organization of the inline content with only minor structure gaps, fitting the 'good structure, minor gaps' anchor.	4 / 5
	Total	15 / 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 states a clear purpose and concrete actions but lacks an explicit 'Use when' trigger and uses second-person voice. It is distinct as a niche but middling on trigger coverage and completeness.

Suggestions

Rewrite in third person and add an explicit trigger clause, e.g. 'Reads, sends, replies to, and checks delivery of email via the assigned AgentMail inbox. Use when working with the assigned AgentMail inbox to handle email tasks.'

Add natural user phrasings and synonyms such as 'email', 'mail', 'message', 'inbox', and 'check delivery' to broaden trigger coverage.

Replace 'Provided automatically by your inbox assignment' with a concrete when-to-use condition so both 'what' and 'when' are explicitly answered.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions ('read email tasks, explicitly send or reply, and check delivery'), which sits at anchor 4, but the second-person imperative 'Use your assigned AgentMail inbox' triggers the voice penalty reducing specificity by one to 3.	3 / 5
Completeness	The 'what' is clear (read/send/reply/check delivery), but there is no 'Use when...' trigger clause; 'Provided automatically by your inbox assignment' is not a when-condition, so the missing-trigger cap of 3 applies.	3 / 5
Trigger Term Quality	Includes relevant keywords (email, inbox, send, reply, delivery) but misses common synonyms and natural phrasings users would say ('mail', 'message', 'check my inbox'), matching the 'some relevant keywords but missing variations' anchor.	3 / 5
Distinctiveness Conflict Risk	'Assigned AgentMail inbox' carves a distinct niche unlikely to fire for unrelated skills; only minor overlap risk with generic mail/message skills, matching the 'mostly distinct' anchor rather than the fully distinct 5.	4 / 5
	Total	13 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: paperclipai/paperclip
Path: skills/agentmail/SKILL.md
Commit: 1c07b59

Reviewed: 1 day 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.