session-sunset

Authoritative protocol for gracefully terminating an agent session. Mandates structured handover comments, mental-model states, and memory persistence to prevent Zero-State Amnesia. Triggers: When concluding a long-running session, executing the Sunset Protocol, handing over work for the next agent, or terminating an agent cycle.

Quality

78%

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 ./.agents/skills/session-sunset/SKILL.md

Quality

Content

72%Weight 40%Scale 1-3

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

The body is a tight, well-structured overview that correctly defers the detailed procedure to a single real reference file. Its weakness is that the actionable sunset steps and validation/feedback loops are entirely delegated, leaving the body itself with decision logic but no in-place procedure.

Suggestions

Inline a minimal sunset checklist (e.g. write handover comment → record mental-model state → persist memory → confirm) so the body is actionable even before the reference is read.

Add an explicit validation/verification step in the body (e.g. "verify the handover note and memory file were written before terminating") to introduce a feedback loop for this destructive operation.

Confirm the deployed reference path matches the bundle layout, since the body cites `.agents/skills/session-sunset/references/...` while the bundle ships `references/session-sunset-workflow.md`.

Dimension	Reasoning	Score
Conciseness	The body is lean bold callouts with no over-explanation of concepts Claude already knows, matching the score-3 "every token earns its place" anchor rather than the padded score-2 anchor.	3 / 3
Actionability	It gives concrete gating conditions ("Context Window Exhaustion (>75% full/forgetfulness)") and an explicit tool/file instruction ("use the `view_file` tool to read ...session-sunset-workflow.md"), but the actual sunset procedure is delegated to the reference rather than present as executable steps here, fitting the score-2 "some concrete guidance but incomplete" anchor rather than the fully copy-paste-ready score 3.	2 / 3
Workflow Clarity	A clear decision sequence exists (anti-trigger halt-gates then valid-condition check) but there is no feedback/validation loop in the body and the multi-step procedure lives in the referenced file; for a risky session-termination op missing feedback loops caps clarity at 2 rather than reaching the score-3 anchor.	2 / 3
Progressive Disclosure	The body is a concise overview that clearly signals one one-level-deep reference ("read ...references/session-sunset-workflow.md"), and that file exists in references/, matching the score-3 anchor for a well-signaled single-level reference rather than the inline-or-implicit score 2.	3 / 3
	Total	10 / 12 Passed

Description

85%Weight 40%Scale 1-3

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, specific description that explicitly covers both what it does and when to use it, with a clear niche. Its main weakness is trigger-term quality: several triggers use internal protocol jargon rather than language a user would naturally utter.

Suggestions

Add user-natural trigger phrasing alongside the protocol terms, e.g. "Use when the context window is filling up, when handing off to another agent, or when the user asks to wrap up the session."

Soften jargon like "executing the Sunset Protocol" and "terminating an agent cycle" with plainer equivalents such as "wrapping up a session" and "ending an agent run."

Dimension	Reasoning	Score
Specificity	Quotes "Mandates structured handover comments, mental-model states, and memory persistence" — it lists multiple specific concrete actions, matching the score-3 anchor rather than the domain-only score-2 anchor.	3 / 3
Completeness	It states what the skill does ("gracefully terminating an agent session", handover/memory persistence) and an explicit when ("Triggers: When concluding a long-running session..."), satisfying the score-3 anchor for both what AND when rather than the implied-when score 2.	3 / 3
Trigger Term Quality	Triggers like "executing the Sunset Protocol" and "terminating an agent cycle" are relevant but lean on protocol jargon rather than terms a user would naturally say, so it fits the score-2 anchor (some relevant keywords, missing common variations) and not the broad natural-coverage of score 3.	2 / 3
Distinctiveness Conflict Risk	The session-termination/handover niche with distinct triggers ("Sunset Protocol", "Zero-State Amnesia") is unlikely to fire for unrelated skills, matching the score-3 clear-niche anchor rather than the overlapping score 2.	3 / 3
	Total	11 / 12 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: neomjs/neo
Path: .agents/skills/session-sunset/SKILL.md
Commit: c1ce853

Reviewed: 8 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.