CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

agentlas-one

Use when the user types /agentlas-one on|off, /agentlas one on|off, or asks to enable/disable the persistent personal agent mode.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

80%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 tight, well-organized command reference with excellent token efficiency and no bundle bloat. Its gaps are operational: no example invocations of the runner, no quick-start sequence tying commands together, and no verification step around the destructive uninstall path.

Suggestions

Add a minimal quick-start sequence (e.g., install → on <name> → status) so the command list becomes a coherent workflow with an implicit verification step.

Show one example invocation of `bin/agentlas-one` (e.g., `bin/agentlas-one on nova`) so the command syntax is copy-paste executable.

For `uninstall [--purge]`, state a verification step after the restore archive is written (e.g., confirm via `status`) to close the destructive-operation validation gap.

DimensionReasoningScore

Conciseness

The body is a lean command reference with zero padding and no explanation of concepts Claude already knows; even the guardrail 'Anything not listed here does not exist — do not invent one' earns its tokens by constraining hallucinated commands — matching anchor 5 ('lean and efficient; every token earns its place').

5 / 5

Actionability

Commands are concrete and real ("status [--runtimes|--drift]", "remember <text>", "orch <sub>") with accurate flags and arguments, but no example invocations of `bin/agentlas-one` are shown and several one-line descriptions (e.g., "seed / memory / curate — the drawer") are too terse to execute confidently — anchor 4 ('mostly executable guidance; concrete commands with minor gaps').

4 / 5

Workflow Clarity

The commands are listed but no sequence is given (e.g., install before on, or how status verifies the install), and the destructive `uninstall [--purge]` has a safety note (a restore archive is written first) but no verification checkpoint — per the destructive-operation guideline, workflow clarity is capped at 3 even though the core on/off action is unambiguous.

3 / 5

Progressive Disclosure

The body is under 50 lines, has no bundle files, and needs none — a well-organized intro plus a Commands section fully covers the skill; per the simple-skill guideline this matches anchor 5.

5 / 5

Total

17

/

20

Passed

Description

73%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 concise, third-person description with explicit and well-anchored trigger conditions and strong distinctiveness. Its main weakness is a thin 'what': it communicates the on/off toggle but not the broader capability surface (install, rename, status, memory curation) that the body documents.

Suggestions

Add 1-2 concrete capability phrases to the description (e.g., install/rename the agent, check status and drift) so the 'what' matches anchor-5 completeness instead of just naming the mode.

Include natural-language synonyms users might say, such as 'turn on/off', 'activate', or 'agent name', alongside the literal slash-command triggers.

Briefly gloss what 'persistent personal agent mode' means (e.g., 'a named agent that persists across sessions') so the domain term is not opaque.

DimensionReasoningScore

Specificity

"enable/disable the persistent personal agent mode" names the domain and the two concrete actions with literal command forms, but the description omits most of the skill's real capabilities (install, rename, status, remember, coverage) documented in the body — matching anchor 3 ('domain and 1-2 concrete actions, but not comprehensive') rather than 4, which requires several specific actions.

3 / 5

Completeness

It has an explicit, concrete 'when' ("Use when the user types /agentlas-one on|off...") and a stated 'what' (enable/disable the persistent personal agent mode), but the 'what' is opaque — the mode itself is unexplained and the action list is thin compared to anchor 5's concrete, multi-action 'what'.

4 / 5

Trigger Term Quality

The description covers both command spelling variants ("/agentlas-one on|off", "/agentlas one on|off") and the natural synonyms "enable/disable", but misses common phrasings like "turn on", "activate", or "agent name" — good coverage with a few natural terms missing (anchor 4).

4 / 5

Distinctiveness Conflict Risk

"agentlas-one" and "persistent personal agent mode" form a clear niche with unique literal trigger phrases (the exact slash commands); there is minimal risk of this firing for a different skill.

5 / 5

Total

16

/

20

Passed

Validation

100%

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

agentlas-ai/Agentlas-OS

Path

.agents/skills/agentlas-one/SKILL.md

Reviewed

Table of Contents

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.