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 ./.claude/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 body excels at workflow sequencing and validation — explicit checklists, exit conditions, and a mechanically verified metric mode with rollback. Its main weaknesses are verbosity from duplicated safety content and display templates, and a monolithic structure that inlines material that should live in separate reference files.

Suggestions

Cut the body to a core process overview plus the metric-mode contract, and move the four worked Patterns and the Self-Regulation weighting tables into a references/ file (e.g. PATTERNS.md, SELF-REGULATION.md) linked from SKILL.md.

Eliminate the duplication between Self-Regulation, Safety Mechanisms, Best Practices, and the Red Flags table by consolidating into a single Safety section stated once.

Replace the display-format markdown templates with brief instructions on what each iteration summary must contain (iteration count, self-regulation score, next change), trusting Claude to format the output.

Dimension	Reasoning	Score
Conciseness	The ~700-line body is noticeably verbose: four fully worked Pattern examples with sample transcripts, markdown display templates Claude does not need spelled out, and the same safety content (max iterations, stall detection, stop-and-ask) restated across Self-Regulation, Safety Mechanisms, Best Practices, and the Red Flags table. Several padded sections match anchor 2 rather than the mostly-efficient profile of anchor 3.	2 / 5
Actionability	Metric Verification Mode is fully executable ("git add -A && git commit -m 'experiment: ...'", "git revert HEAD --no-edit", "mkdir -p .claude-octopus/experiments", exact JSONL fields), but the standard-loop half relies on placeholder templates ("[what to do each loop]", "[Action 1] → [result]") and vague mental stall-detection. Falls between anchor 3 (pseudocode/incomplete) and anchor 4 (mostly executable, minor gaps).	3.5 / 5
Workflow Clarity	The phases are clearly sequenced (Setup → Execution → Exit Conditions), validation is explicit (Safety Validation checklist, three distinct exit conditions with user options, guard commands, revert-on-regression feedback loop, resume-from-log behavior). This matches anchor 5: explicit validation steps, feedback loops for error recovery, and checklists.	5 / 5
Progressive Disclosure	No bundle files exist (no references/, scripts/, or assets/ directories), and the body inlines ~150 lines of Metric Verification Mode plus four worked patterns that belong in reference files. Section headers give some structure, matching anchor 3 ('content that should be separate is inline') rather than anchor 4's appropriate split.	3 / 5
	Total	13.5 / 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 clearly states what the skill does and when to use it, with several natural trigger terms. It is held back by thin action specificity, missing user-phrase synonyms like "retry" and "keep trying", and moderate overlap risk with adjacent debugging/retry skills.

Suggestions

Add concrete user trigger phrases, e.g. "Use when the user says 'loop N times', 'keep trying until', 'retry up to N times', or 'iterate until it passes'."

Name one or two concrete capabilities, e.g. "...with max-iteration limits, progress tracking, and automatic rollback on regression" to sharpen specificity and distinctiveness.

Distinguish the skill from debugging/retry skills in the description, e.g. "Use this instead of skill-debug when multiple bounded iterations are requested."

Dimension	Reasoning	Score
Specificity	"Run tasks in a loop until goals are met" names the domain and one core action, and "iterative refinement, polling, or convergence" lists use modes, but no concrete mechanics (max iterations, exit conditions, metric verification) appear. Matches 'names domain and 1-2 concrete actions, but not comprehensive' rather than the fuller action list of anchor 4.	3 / 5
Completeness	Both parts are explicit: "Run tasks in a loop until goals are met" (what) and "use for iterative refinement, polling, or convergence" (when). The 'when' names categories rather than concrete user trigger phrases, so it fits anchor 4 ('when' could be more explicit or specific) instead of anchor 5.	4 / 5
Trigger Term Quality	Natural terms like "loop", "until goals are met", "iterative", and "polling" are present, but common user variations such as "retry", "repeat", "keep trying until", and "iterate N times" are missing. Good coverage with a few natural terms absent fits anchor 4, not the synonym-complete coverage of anchor 5.	4 / 5
Distinctiveness Conflict Risk	Looping is a somewhat generic mechanism that overlaps with debugging, retry, and iterative code-improvement skills; "polling" additionally overlaps monitoring/automation skills. Matches 'somewhat specific but could still overlap with similar skills' rather than the minor-overlap profile of anchor 4.	3 / 5
	Total	14 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (726 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: nyldn/claude-octopus
Path: .claude/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.