onboard

Onboard a new user OR curate a single piece of the assistant's inner state. Without arguments, runs the full first-run ceremony (AI name, personality, user profile, SOUL.md + USER.md, optional browser + personal website). With `scope:soul` or `scope:user`, runs a quick chat to update just that one profile file. With `path:<abs>`, runs a quick chat to update / keep / delete one memory file under ~/.clacky/memories/.

Quality

73%

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 ./lib/clacky/default_skills/onboard/SKILL.md

Quality

Content

71%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 thorough, highly actionable bilingual onboarding skill with concrete cards, templates, and commands, strong sequencing, and proper confirmations around destructive actions. Its weaknesses are token weight from full zh/en duplication and inlined bulk content that would benefit from reference files given the body's length.

Suggestions

Move the large bilingual card/template library and the A.13 closing-moment prose into one-level-deep reference files (e.g. `references/cards.json`, `references/templates.md`) and link them from the body to improve progressive disclosure and cut inline tokens.

Add an explicit validate-after-write checkpoint for SOUL.md/USER.md (e.g. re-read the file to confirm structure and language) to lift workflow clarity above 4.

Where zh and en cards differ only by translation, consider a single parameterized card with a language switch to reduce duplicated JSON blocks.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with no concept-over-explanation padding, but the full bilingual (zh+en) duplication of every card/template and the lengthy emotional A.13 closing blocks (three variants × two languages) consume tokens that could be tightened or externalized.	3 / 5
Actionability	Copy-paste ready throughout: exact `ask_user` JSON cards, concrete file paths, full SOUL.md/USER.md templates, and specific shell commands (`test -d ~/.openclaw`, the WSL `powershell.exe` probe) plus explicit `trash`-tool usage for deletion.	5 / 5
Workflow Clarity	Clear three-mode dispatch with numbered A.1–A.14 / B.1–B.8 / C.1–C.8 sequences and confirmation checkpoints (B.7 confirm card, C.6c delete re-confirm), but no explicit validate-after-write step for SOUL.md/USER.md and checkpoints are mostly ask_user cards rather than verification commands.	4 / 5
Progressive Disclosure	Sections are well organized and the two helper scripts are correctly externalized, but the ~570-line body inlines the entire bilingual card/template library and closing-moment prose that, given the size, would navigate better as one-level-deep reference files (e.g. cards.md, templates/).	3 / 5
	Total	15 / 20 Passed

Description

75%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 well-specified multi-mode skill description that maps explicit invocation arguments to concrete behaviors and outcomes. Its main weakness is the absence of a plain-language "Use when..." trigger phrase, relying instead on argument syntax to convey when to fire.

Suggestions

Add a natural-language trigger clause (e.g. "Use when a new user is being onboarded or when tweaking the assistant's SOUL/USER profile or a single memory file") so the skill fires on plain user intent, not just argument syntax.

Surface the default fallbacks (AI name "Clacky", user "Boss"/"老大") only if they aid discoverability; otherwise keep them in the body to avoid diluting the description.

Consider a synonym like "set up"/"personalize" alongside "onboard" to catch phrasings users actually say.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions across three modes ("name the AI, pick personality", "write SOUL.md + USER.md", "update / keep / delete one memory file") with only minor gaps in coverage.	4 / 5
Completeness	Clearly answers "what" (per-mode outcomes) and "when" via explicit invocation conditions ("Without arguments", "With `scope:soul`", "With `path:<abs>`"), but lacks a natural-language "Use when..." clause a user would voice.	4 / 5
Trigger Term Quality	Includes natural triggers like "Onboard a new user", "curate", plus arg tokens `scope:soul`, `scope:user`, `path:<abs>`; a few common synonyms a user might naturally say are missing.	4 / 5
Distinctiveness Conflict Risk	Has a clear onboarding/identity-curation niche with arg-gated dispatch that minimizes misfires; minor overlap risk with the memory-curation mode and the skills it merely invokes (browser-setup, personal-website).	4 / 5
	Total	16 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (579 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: clacky-ai/openclacky
Path: lib/clacky/default_skills/onboard/SKILL.md
Commit: db23154

Reviewed: 2 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.