autonomous-agent-harness

Transform Claude Code into a fully autonomous agent system with persistent memory, scheduled operations, computer use, and task queuing. Replaces standalone agent frameworks (Hermes, AutoGPT) by leveraging Claude Code's native crons, dispatch, MCP tools, and memory. Use when the user wants continuous autonomous operation, scheduled tasks, or a self-directing agent loop.

Quality

67%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/autonomous-agent-harness/SKILL.md

Quality

Content

50%

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

A well-structured, concrete skill body with useful cron/dispatch examples, but it is held back by conceptual padding, non-executable package/endpoint names, missing inline validation loops, and a monolithic layout that would benefit from reference files.

Suggestions

Verify and correct the MCP package names and the dispatch endpoint so the setup commands are actually copy-paste executable (e.g. use the real @modelcontextprotocol/server-memory package rather than @anthropic/memory-mcp-server).

Embed explicit validate→fix→retry checkpoints inside the example workflows (e.g. confirm a cron task ran, check its output, and retry/alert on failure) instead of only mentioning error handling as a constraint.

Move the Hermes comparison table, setup guide, and example workflows into one-level-deep reference files (e.g. MIGRATION.md, SETUP.md, WORKFLOWS.md) referenced from a leaner SKILL.md overview to improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with concrete tables and commands, but it explains concepts Claude already knows (e.g. built-in memory storing 'User preferences, feedback, project context' as markdown with frontmatter) and includes a large decorative ASCII architecture diagram that could be trimmed.	2 / 3
Actionability	It provides specific cron expressions, JSON config, curl/claude -p commands, and MCP call examples, but copy-paste readiness is undermined by non-existent package names ('@anthropic/memory-mcp-server') and a fabricated dispatch endpoint, leaving key details unverifiable.	2 / 3
Workflow Clarity	The setup guide and example workflows are clearly sequenced with numbered steps, but for scheduled/batch operations the rubric caps at 2 when feedback loops are missing — validation is only mentioned as a constraint ('Add error handling to cron prompts') rather than embedded as validate→fix→retry checkpoints inside the workflows.	2 / 3
Progressive Disclosure	Sections are well organized, but the ~275-line SKILL.md is monolithic: the Hermes comparison table, setup guide, and example workflows are all inline with no one-level-deep reference files, so content that should be split out is not.	2 / 3
	Total	8 / 12 Passed

Description

85%

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, third-person description that states concrete capabilities and an explicit 'Use when' trigger, with a clear distinct niche. Its only gap is that the most natural user phrasings sit in the body rather than the description field itself.

Dimension	Reasoning	Score
Specificity	Names multiple concrete capabilities — 'persistent memory, scheduled operations, computer use, and task queuing' — and ties them to native primitives (crons, dispatch, MCP tools), matching the anchor that lists several specific concrete actions.	3 / 3
Completeness	Explicitly answers both what ('Transform Claude Code into a fully autonomous agent system with...') and when ('Use when the user wants continuous autonomous operation, scheduled tasks, or a self-directing agent loop.'), matching the anchor with explicit triggers.	3 / 3
Trigger Term Quality	It has a 'Use when...' clause with relevant terms ('scheduled tasks', 'continuous autonomous operation', 'self-directing agent loop', Hermes/AutoGPT), but the most natural user phrasings ('run this every day', 'keep monitoring') live in the body, not the description, and several terms are somewhat jargon-heavy.	2 / 3
Distinctiveness Conflict Risk	It carves a clear niche — an autonomous-agent harness replacing Hermes/AutoGPT via Claude Code native features — with distinct triggers unlikely to fire for unrelated skills.	3 / 3
	Total	11 / 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
metadata_version	'metadata.version' is missing	Warning

	Total	15 / 16 Passed

Repository: affaan-m/ECC
Path: skills/autonomous-agent-harness/SKILL.md
Commit: 754b8dd

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.