skill-iterative-loop

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

Quality

65%

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 ./.claude/skills/skill-iterative-loop/SKILL.md

Quality

Content

62%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A thorough, well-sequenced loop-execution skill with excellent workflow clarity and a genuinely executable metric-verification mode. Its weaknesses are verbosity from redundant safety sections and placeholder templates, plus a monolithic structure that does not split large sections into reference files.

Suggestions

Consolidate the overlapping "Best Practices", "Red Flags", and "Safety Mechanisms" sections into one to reduce redundancy and trim the emoji-heavy exit templates.

Replace placeholder templates in the core loop phases with a concrete, fully worked walkthrough (commands and expected outputs) so the primary path is as actionable as the metric mode.

Move the large Metric Verification Mode and Self-Regulation sections into separate reference files (e.g., METRIC_MODE.md, SELF_REGULATION.md) referenced from SKILL.md to improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body is accurate and lean in places, but it is padded with emoji-laden exit-condition templates, placeholder fill-in blocks, and redundant coverage of the same safety ideas across "Best Practices", "Red Flags", and "Safety Mechanisms".	2 / 3
Actionability	The Metric Verification Mode is fully executable (exact git/JSONL commands, parameter table, 9-step contract), but the core Phase 1–3 loop process relies on placeholder templates like "[what should be achieved]" and "[N]" rather than complete, copy-paste-ready guidance.	2 / 3
Workflow Clarity	The skill is explicitly sequenced into Setup, Execution, and Exit phases with validation checkpoints ("Never proceed without max iterations defined", WTF-score thresholds, revert-on-regression) and clear feedback loops for error recovery.	3 / 3
Progressive Disclosure	The content is well-sectioned, but it is a ~700-line monolithic SKILL.md with no bundle files; substantial sections like Metric Verification Mode and Self-Regulation could be split into separate reference files, and the only external reference (codex-host-adapter.md) is not part of this skill's bundle.	2 / 3
	Total	9 / 12 Passed

Description

67%

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 that covers both capability and explicit trigger guidance, with a natural keyword ("loop"). It stops short of listing concrete operations and is slightly broad on the trigger side, capping specificity and distinctiveness at the middle level.

Suggestions

Add concrete actions or example verbs (e.g., "retry failed operations, optimize a metric over rounds, converge on a passing test suite") to lift specificity toward 3.

Broaden trigger terms with everyday phrasings users actually say ("retry", "repeat until", "keep trying until") alongside the technical terms.

Sharpen distinctiveness by contrasting with adjacent skills in the description itself (e.g., "for bounded multi-round iteration, not single retries or one-off debug passes").

Dimension	Reasoning	Score
Specificity	"Run tasks in a loop until goals are met" names the action and domain, and "iterative refinement, polling, or convergence" lists use-case categories, but these are abstract categories rather than multiple concrete operations, so it does not reach the level of listing several specific concrete actions.	2 / 3
Completeness	It states what the skill does ("Run tasks in a loop until goals are met") and provides an explicit trigger clause ("use for iterative refinement, polling, or convergence"), answering both what and when.	3 / 3
Trigger Term Quality	"loop" is a natural term a user would say, and "iterative refinement" is reasonable, but "polling" and "convergence" lean technical and common variations such as "retry", "repeat", or "keep trying until" are missing.	2 / 3
Distinctiveness Conflict Risk	Loop execution with goals/polling/convergence is a recognizable niche, but "iterative refinement" is broad and could overlap with retry or debugging skills, so it is not fully conflict-free.	2 / 3
	Total	9 / 12 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 (699 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: nyldn/claude-octopus
Path: .claude/skills/skill-iterative-loop/SKILL.md
Commit: 9f39f0c

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.