skill-iterative-loop

Run tasks in a loop until goals are met — use for iterative refinement, polling, or convergence

Quality

62%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/skill-iterative-loop/SKILL.md

Quality

Content

58%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.

The content is highly actionable with an exceptionally clear, validated workflow (the metric-verification execution contract is the strongest part), but it is held back by significant verbosity from duplicated sections and by a monolithic structure that inlines content which belongs in separate reference files, including one dangling reference to a missing file.

Suggestions

Split Metric Verification Mode and Common Patterns into reference files (e.g. references/metric-mode.md, references/patterns.md), keeping SKILL.md as an overview with clearly signaled one-level-deep links.

Remove the duplicated restatements: 'Safety Mechanisms' items 2-4 restate Self-Regulation, 'Best Practices' restates the Phase 3 exit conditions, and Strategy Rotation is covered twice.

Delete summary padding ('The Bottom Line', 'Quick Reference' table) and fix or remove the dangling 'skills/blocks/codex-host-adapter.md' reference, plus the '/octo:loop' example command the host note says is unavailable.

Dimension	Reasoning	Score
Conciseness	The ~700-line body is noticeably verbose with several padded sections: 'The Bottom Line' summary, four long Common Patterns, and content stated twice (Self-Regulation details restated in 'Safety Mechanisms', exit conditions restated in 'Best Practices', strategy rotation covered in two places). Not a 1 because it does not explain concepts Claude already knows — the bulk is operational guidance.	2 / 5
Actionability	Mostly executable guidance: copy-paste-ready commands ('git add -A && git commit -m "experiment: ..."', 'git revert HEAD --no-edit'), exact JSONL field specs, a conf file format, and a 9-step execution contract. Minor gaps: the '/octo:loop' example uses a slash command the host note says does not exist in Codex, host-specific tools like 'AskUserQuestion' are assumed, and some policy blocks are pseudocode.	4 / 5
Workflow Clarity	Clear sequences with explicit validation at every risky step: a safety checklist before starting ('Never proceed without max iterations defined'), metric and guard verification before keeping changes, automatic revert on regression, resume-from-log behavior, and stall-detection feedback loops. Destructive operations (git revert) are always guarded, so the destructive-operation cap does not apply.	5 / 5
Progressive Disclosure	A ~700-line monolithic file with no bundle files, yet the body references 'skills/blocks/codex-host-adapter.md' which does not exist in this bundle. Major self-contained sections (the ~150-line Metric Verification Mode, Common Patterns) that clearly belong in separate reference files are fully inlined.	2 / 5
	Total	13 / 20 Passed

Description

66%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 with a proper third-person 'use for' clause and reasonable trigger terms. Its weaknesses are moderate: trigger coverage lacks common synonyms (retry, keep trying until), and the phrasing is broad enough to overlap with sibling debugging/retry skills.

Suggestions

Add natural trigger synonyms users actually say, e.g. 'use when the user asks to loop, retry, keep trying, or iterate until something passes or converges'.

Make the 'when' clause more situation-specific (e.g. 'until tests pass', 'until a metric hits a target') to distinguish it from debugging and generic retry skills.

Mention the metric-verification mode in the description so users searching for 'optimize until < 100ms' or 'improve coverage' find this skill.

Dimension	Reasoning	Score
Specificity	The description names the domain (iterative loop execution) with the concrete action 'Run tasks in a loop until goals are met' plus 'iterative refinement, polling, or convergence' — domain plus 1-2 concrete actions, but not a comprehensive list of specific actions.	3 / 5
Completeness	It has a clear what ('Run tasks in a loop until goals are met') and an explicit when ('use for iterative refinement, polling, or convergence'). The when clause names modes rather than concrete user-situation trigger phrases, keeping it below a 5.	4 / 5
Trigger Term Quality	'loop', 'iterative refinement', 'polling', 'convergence' are natural phrases users would say. Missing common synonyms such as 'retry', 'keep trying until', or 'iterate until it works' — patterns the body itself covers.	4 / 5
Distinctiveness Conflict Risk	'Run tasks in a loop until goals are met' is broad and would overlap with debugging, retry, and general agent-loop skills (the body itself cross-references skill-debug for retry logic), so overlap risk is more than minor.	3 / 5
	Total	14 / 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 (700 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: nyldn/claude-octopus
Path: skills/skill-iterative-loop/SKILL.md
Commit: 0093dab

Reviewed: 5 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.