CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ar-loop

Use for /ar-loop, or to run an autoresearch experiment on a schedule (10 min to monthly) with CronCreate.

Quality

74%

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

Fix and improve this skill with Tessl

tessl review fix ./engineering/autoresearch-agent/skills/ar-loop/SKILL.md
SKILL.md
Quality
Evals
Security

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 well-structured, highly actionable command skill: the cron mapping, prompt template, and metadata schema are concrete and the loop includes explicit evaluation and crash-escalation feedback. Remaining gaps are minor: small redundancy around the 3-day expiry, no setup-phase verification, and a referenced script that is absent from the bundle.

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence ('Use CronCreate with this prompt', 'Call CronDelete with that ID') with no explanation of known concepts, but has minor duplication: the 3-day auto-expiry note appears both in the Step 5 confirmation block ('Recurring jobs auto-expire after 3 days') and again in 'Important Limitations', and the Step 2 interval menu repeats the cron mapping table.

4 / 5

Actionability

Provides exact cron expressions, a fill-in CronCreate prompt template, an exact loop.json schema, and a concrete stop procedure, but references '{skill_path}/scripts/run_experiment.py' which is not present in the bundle (no scripts/ directory), leaving a minor gap in full executability.

4 / 5

Workflow Clarity

Steps 1-5 are clearly sequenced and the recurring loop embeds a feedback loop ('Read the output (KEEP/DISCARD/CRASH)') with error escalation ('If 5 consecutive crashes... delete this cron job (CronDelete) and alert'), but setup lacks verification checkpoints (e.g., confirming CronCreate succeeded or that the experiment branch and results.tsv exist before starting).

4 / 5

Progressive Disclosure

As a single-file skill with no bundle files, it is well-sectioned (Usage, What It Does, Stopping a Loop, Important Limitations) and its one referenced path (scripts/run_experiment.py) is clearly signaled and one level deep, though that file does not exist in the bundle; content is appropriately inline for a skill of this size.

4 / 5

Total

16

/

20

Passed

Description

62%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.

The description answers both what and when explicitly and sits in a distinct niche thanks to the /ar-loop command and 'autoresearch' domain. Its weaknesses are limited action coverage and missing natural trigger synonyms like 'loop', 'recurring', or 'cron job'.

Suggestions

Add natural trigger synonyms users would actually say — e.g., 'recurring', 'loop', 'periodically', or 'cron job' — so the description fires on phrasings like 'run my experiment on a loop'.

List the concrete lifecycle actions the skill performs (start a recurring experiment loop, stop an active loop, resume after auto-expiry) to raise specificity beyond a single action.

Sharpen the when-clause toward user intent, e.g., 'Use when the user wants an autoresearch experiment to iterate automatically on a schedule (10 min to monthly)'.

DimensionReasoningScore

Specificity

'run an autoresearch experiment on a schedule (10 min to monthly) with CronCreate' names the domain plus one concrete action and mechanism, matching anchor 3 (1-2 concrete actions, not comprehensive); it omits the loop lifecycle actions the body covers (starting and stopping recurring loops, handling auto-expiry), so it is not 4.

3 / 5

Completeness

Both parts are present: a clear 'what' ('run an autoresearch experiment on a schedule... with CronCreate') and an explicit 'when' ('Use for /ar-loop, or to...'), but the when-clause could state the user intent more concretely (e.g., when the user wants automated, repeated experiment iterations), matching anchor 4 rather than 5.

4 / 5

Trigger Term Quality

It includes '/ar-loop', 'experiment', 'schedule', and '10 min to monthly', but misses natural variations a user would say such as 'loop', 'recurring', 'periodically', or 'cron job', and 'CronCreate' is tool jargon rather than a user phrase — matching anchor 3 (relevant keywords but missing common variations).

3 / 5

Distinctiveness Conflict Risk

'autoresearch experiment' plus the unique '/ar-loop' command trigger gives it a clear niche with minimal conflict risk, though 'run... on a schedule' has minor overlap with generic scheduling/cron skills, matching anchor 4 (mostly distinct, minor overlap risk).

4 / 5

Total

14

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository

alirezarezvani/claude-skills

Path

engineering/autoresearch-agent/skills/ar-loop/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.