handoff

Draft a self-contained prompt for a fresh ChatGPT, Claude Code, or other agent session. Use when the user asks for a handoff, continuation prompt, or something to copy and paste into another session. Do not use for a progress summary or for asking another agent to review or implement work within the active session.

Quality

81%

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

75%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 disciplined, example-light instruction skill: the grounding pass gives real executable commands, the workflow is clearly sequenced and ends with an explicit self-validation check, and nothing wastes tokens on concepts Claude already knows. Its weaknesses are a persistent abstract register in the writing-guidance passages (no worked example of a good vs. bad handoff prompt) and two dangling cross-skill references.

Suggestions

Fix or qualify the two cross-skill links: '../adversarial-review/SKILL.md' and '../agent-goal/SKILL.md' do not exist in this bundle, so a reader following them finds nothing — either ship those files, point to where they live, or reword as plain references.

Add one compact worked example (a short good handoff excerpt, or a good-vs-bad pair) in the 'Write it' section to ground the abstract directives like 'Say what you think is true and why' and 'Lead with the mission and the artifact wanted'.

Tighten the flourish-heavy passages (e.g., 'Two things are easy to leave out and expensive to lose...' and the closing of 'The posture that makes it work') into direct imperatives to close the remaining conciseness gap.

Dimension	Reasoning	Score
Conciseness	Dense, high-signal prose that assumes Claude's competence and never explains known concepts (the git block is shown, not taught); a few abstract flourishes ('Two things are easy to leave out and expensive to lose', 'The posture that makes it work') could be trimmed, keeping it below the lean 5 anchor.	4 / 5
Actionability	Provides executable commands ('git status --short --branch', 'git diff --name-status', 'git diff -- <relevant paths>') and concrete directives ('Lead with the mission and the artifact wanted', 'Close with the likely verification commands'), but several passages remain abstract direction ('Say what you think is true and why') without a worked example — minor gaps versus fully executable coverage.	4 / 5
Workflow Clarity	Clear three-stage sequence (ground it first → write it → posture/close) with a terminal validation checkpoint ('Read the prompt as someone arriving cold: can they tell what to accomplish...'), but there is no revise-and-recheck feedback loop if that cold-read fails, so the 5 anchor is not met; checkpoints are explicit, well above level 3.	4 / 5
Progressive Disclosure	Well-organized single-file skill with clear section headers and clearly signaled, one-level-deep links to related skills; however, both referenced paths ('../adversarial-review/SKILL.md', '../agent-goal/SKILL.md') do not resolve within this bundle, which is a reference-integrity gap that keeps it below the 5 anchor.	4 / 5
	Total	16 / 20 Passed

Description

82%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 voice, concrete artifact, explicit 'Use when...' triggers, and unusually good negative boundary guidance that separates it from summary and in-session delegation skills. The only soft spot is specificity — it names a single concrete action rather than enumerating several capabilities.

Dimension	Reasoning	Score
Specificity	Names the domain (handoff/continuation to a fresh agent session) and one concrete action — 'Draft a self-contained prompt' — but does not list several distinct concrete actions; a 4 would require several specific actions with only minor gaps.	3 / 5
Completeness	Explicitly answers 'what' ('Draft a self-contained prompt for a fresh ChatGPT, Claude Code, or other agent session') and 'when' ('Use when the user asks for a handoff, continuation prompt, or something to copy and paste into another session') with concrete trigger phrases, plus negative triggers; a 4 would require a less explicit 'when'.	5 / 5
Trigger Term Quality	Includes natural phrases users would say — 'handoff', 'continuation prompt', 'copy and paste into another session', 'fresh ... session' — with good coverage; a few natural synonyms (e.g., 'hand this off', 'brief the next agent') are missing, so it falls between the 4 and 5 anchors.	4 / 5
Distinctiveness Conflict Risk	Clear niche (prompt artifact for a fresh session rather than in-session work), distinct trigger phrases, and explicit boundary guidance ('Do not use for a progress summary or for asking another agent to review or implement work within the active session') that prevents overlap with adjacent skills — minimal conflict risk.	5 / 5
	Total	17 / 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: 2 suspicious	Warning

	Total	14 / 16 Passed

Repository: EpicenterHQ/epicenter
Path: .agents/skills/handoff/SKILL.md
Commit: f9441c8

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.