autonomous-loops

Patterns and architectures for autonomous Claude Code loops — from simple sequential pipelines to RFC-driven multi-agent DAG systems. Retained for compatibility only: when new autonomous loop guidance is needed, use continuous-agent-loop instead.

Quality

63%

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/autonomous-loops/SKILL.md

Quality

Content

63%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, largely executable catalog of six autonomous-loop patterns with strong workflow sequencing and feedback loops, weakened by verbosity/rationale padding and a monolithic single-file layout that forgoes progressive disclosure into separate reference files.

Suggestions

Move the detailed pattern implementations (e.g., the full `.claude/commands/infinite.md` spec and the Ralphinho architecture/interfaces) into separate reference files under references/ and link to them from a concise overview, improving both conciseness and progressive disclosure.

Trim the 'Core Insight' blockquotes and rationale sections that explain concepts Claude already knows (what `claude -p` does, why negative instructions backfire) to tighten token efficiency.

Add explicit validation/verification checkpoints to the Infinite Agentic Loop and NanoClaw sections so every batch-oriented pattern has a validate-then-proceed gate.

Dimension	Reasoning	Score
Conciseness	The ~600-line body is dense and mostly useful pattern-specific content with executable code, but it includes several padded rationale/'Core Insight' sections and explanations of concepts Claude already knows (e.g., what `claude -p` does, why negative instructions are dangerous), keeping it from 4.	3 / 5
Actionability	Provides copy-paste-ready bash scripts, model-routing flags, a full `.claude/commands/infinite.md` command file, continuous-claude CLI usage, and a TypeScript WorkUnit interface; the most complex pattern (Ralphinho) is more diagram/interface-described than runnable, which is the minor gap preventing 5.	4 / 5
Workflow Clarity	Multi-step processes are clearly sequenced (sequential pipeline steps, continuous-claude iteration 1-9, DAG layers, merge queue) with explicit feedback loops (CI failure recovery, eviction-with-context re-runs, de-sloppify verify-then-commit) and a decision matrix; a couple of patterns (Infinite Agentic Loop, NanoClaw) lack explicit validation checkpoints, holding it at 4.	4 / 5
Progressive Disclosure	Internal sectioning and the pattern-spectrum table give good in-document navigation, but no bundle files exist and all six patterns — including large blocks that would suit separate reference files (the infinite-loop command spec, the full Ralphinho architecture) — are inlined in one monolithic SKILL.md.	3 / 5
	Total	14 / 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 clearly states what the skill covers and names a concrete spectrum of loop patterns, but omits an explicit 'when to use' trigger and self-identifies as a deprecated/compatibility skill that overlaps with continuous-agent-loop.

Suggestions

Add an explicit 'Use when...' clause stating when Claude should reach for this skill (e.g., 'Use when setting up autonomous claude -p pipelines or multi-agent DAG orchestration').

Reconcile the compatibility redirect: either sharpen the distinctiveness from continuous-agent-loop or make the deprecation/trigger boundary explicit so the skill does not fire for the wrong loop request.

Include common user-facing synonyms (e.g., 'CI/CD for Claude', 'continuous development loops') alongside the technical pattern names to broaden trigger coverage.

Dimension	Reasoning	Score
Specificity	Names the domain ('autonomous Claude Code loops') and several concrete pattern categories ('simple sequential pipelines', 'RFC-driven multi-agent DAG systems'), giving comprehensive coverage of the spectrum; stops short of 5 because it lists pattern/architecture categories rather than concrete executable actions.	4 / 5
Completeness	It has a clear 'what' (patterns and architectures for autonomous loops) but no explicit 'Use when...' trigger guidance — the compatibility/redirect clause is a pointer to another skill, not a usage trigger, which per the rubric caps completeness at 3.	3 / 5
Trigger Term Quality	Includes natural terms a user might say ('autonomous Claude Code loops', 'sequential pipelines', 'multi-agent DAG') plus the redirect name 'continuous-agent-loop'; a few common synonyms or phrasings are missing, so it does not reach 5.	4 / 5
Distinctiveness Conflict Risk	The niche ('autonomous Claude Code loops') is reasonably specific, but the explicit 'Retained for compatibility only ... use continuous-agent-loop instead' note signals real overlap with a sibling skill, leaving moderate conflict risk.	3 / 5
	Total	14 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (612 lines); consider splitting into references/ and linking	Warning
metadata_version	'metadata.version' is missing	Warning
referenced_paths_exist	Referenced path issues: 2 missing	Warning

	Total	13 / 16 Passed

Repository: affaan-m/ECC
Path: skills/autonomous-loops/SKILL.md
Commit: d8409a4

Reviewed: 1 day 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.