loop

Start an autonomous experiment loop with user-selected interval (10min, 1h, daily, weekly, monthly). Uses CronCreate for scheduling. Use when the user runs /ar:loop or asks to run an autoresearch experiment continuously on a schedule.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

—

The risk profile of this skill

Fix and improve this skill with Tessl

tessl review fix ./engineering/autoresearch-agent/skills/loop/SKILL.md

Quality

Content

68%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 concrete, well-structured workflow skill with copy-paste-ready cron mappings and a complete scheduling prompt, weakened mainly by missing validation/feedback loops for its batch operations and a referenced script that has no corresponding bundle file.

Suggestions

Add an explicit validate→fix→retry checkpoint after the evaluate step (e.g., interpret KEEP/DISCARD/CRASH, and on DISCARD or CRASH roll back the commit before the next iteration) to lift workflow clarity above the batch-operation cap.

Clarify the KEEP/DISCARD/CRASH output format and what action each verdict triggers, since the prompt references it but never defines it.

Either provide the referenced scripts/run_experiment.py in a scripts/ bundle or inline its contract so the run command is genuinely executable.

Dimension	Reasoning	Score
Conciseness	The body is efficient — tables for cron mappings, a filled-in CronCreate prompt template, and concise numbered steps — with only minor trim opportunities such as the 3-day auto-expiry note repeated in both the confirm output and the Limitations section.	4 / 5
Actionability	Mostly executable: concrete cron expressions, a complete CronCreate prompt template, explicit file paths (.autoresearch/{domain}/{name}/...), and real git/python commands; minor gaps like 'list experiments and let user pick' and the unexplained KEEP/DISCARD/CRASH output keep it from a 5.	4 / 5
Workflow Clarity	Steps 1–5 are clearly sequenced and the crash-threshold rule ('If 5 consecutive crashes... delete this cron job and alert') is a feedback checkpoint, but this batch/destructive skill (automated commits, experiment runs, cron deletion) lacks a general validate→fix→retry loop, capping it at 3 per the feedback-loop guideline.	3 / 5
Progressive Disclosure	The body is well-organized into clear sections (Usage, What It Does, Stopping, Limitations) with no nested references; the only external reference, 'python {skill_path}/scripts/run_experiment.py', points to a scripts/ directory that does not exist, a minor organization gap.	4 / 5
	Total	15 / 20 Passed

Description

87%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, specific description that clearly states both what the skill does and when to invoke it, with concrete interval options and a distinct autoresearch niche. Minor room to broaden trigger synonyms beyond the autoresearch-specific vocabulary.

Dimension	Reasoning	Score
Specificity	Quotes 'Start an autonomous experiment loop with user-selected interval (10min, 1h, daily, weekly, monthly). Uses CronCreate for scheduling' — names the domain and several concrete capabilities with explicit interval options, but stops short of comprehensively listing what the loop does inside each run.	4 / 5
Completeness	Both halves are explicit: the 'what' ('Start an autonomous experiment loop with user-selected interval... Uses CronCreate for scheduling') and the 'when' ('Use when the user runs /ar:loop or asks to run an autoresearch experiment continuously on a schedule') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Trigger phrase 'Use when the user runs /ar:loop or asks to run an autoresearch experiment continuously on a schedule' gives good natural coverage (experiment, loop, continuously, schedule, the command), though terms skew toward the autoresearch jargon rather than broad synonyms.	4 / 5
Distinctiveness Conflict Risk	'autonomous experiment loop', 'autoresearch experiment continuously on a schedule', and the /ar:loop command carve a clear niche with distinct triggers and minimal overlap with other skills.	5 / 5
	Total	18 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: alirezarezvani/claude-skills
Path: engineering/autoresearch-agent/skills/loop/SKILL.md
Commit: 19392f7

Reviewed: ~7 hours 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.