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

83%

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

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.

A well-architected progressive-disclosure skill: the SKILL.md body is a lean overview with a clear validation gate that defers detail to a single reference file. Weakest in actionability and progressive_disclosure only because the real workflow and bulk content live in the monolithic reference file.

Suggestions

Use a markdown link for the reference path (e.g., [session-sunset-workflow.md](references/session-sunset-workflow.md)) instead of a raw filesystem string for clearer navigation.

Consider splitting the 226-line reference into focused sub-files (e.g., trigger-conditions, handoff-steps, payload-template) so the body can link to each, reducing the monolithic reference.

Inline a one-line summary of the 10-step handoff sequence in the body so the core workflow is visible without opening the reference.

Dimension	Reasoning	Score
Conciseness	The body is lean and efficient: a compact anti-trigger guard plus a single concrete pointer to the reference file, with no padding or explanation of concepts Claude already knows.	5 / 5
Actionability	Gives one fully concrete, executable instruction ('use the view_file tool to read .agents/skills/session-sunset/references/session-sunset-workflow.md'), but the actual multi-step workflow is deferred to the reference rather than present in the body.	4 / 5
Workflow Clarity	Clear sequenced gate (check anti-triggers -> confirm valid sunset condition -> read reference -> adhere before terminating) with an explicit validation checkpoint (the Completion Bias Guard), though detailed validation lives in the deferred reference.	4 / 5
Progressive Disclosure	Good structure: the body is an overview pointing to one real, clearly-signaled one-level-deep reference (verified present at references/session-sunset-workflow.md), but the reference is a dense monolithic file referenced via a raw path rather than a markdown link.	4 / 5
	Total	17 / 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, well-structured description that clearly states both capability and explicit trigger conditions, with concrete mandated actions and a distinctive niche. The main weakness is protocol-level abstraction in the action list and minor overlap risk with general handover skills.

Dimension	Reasoning	Score
Specificity	Names the domain (terminating an agent session) and several concrete mandated actions ('structured handover comments, mental-model states, and memory persistence'), though the actions are protocol-level rather than tool-level, leaving minor coverage gaps.	4 / 5
Completeness	Explicitly answers both 'what' (authoritative protocol for terminating an agent session mandating handover comments, mental-model states, memory persistence) and 'when' via a concrete 'Triggers:' clause with multiple trigger phrases.	5 / 5
Trigger Term Quality	Lists several natural trigger phrases ('concluding a long-running session', 'handing over work for the next agent', 'terminating an agent cycle') with good coverage, but lacks synonyms or variations that would push it to comprehensive.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (session-termination handover) with distinctive named concepts ('Zero-State Amnesia', 'Sunset Protocol'), but 'handing over work for the next agent' carries minor overlap risk with generic handover/workflow skills.	4 / 5
	Total	17 / 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: neomjs/neo
Path: .agents/skills/session-sunset/SKILL.md
Commit: 210c03c

Reviewed: 4 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.