CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

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

77%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

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

Quality

Content

61%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 delivers genuinely actionable setup guidance with concrete commands, a pinned MCP config, and a useful cron-pattern table, but it is padded by a large decorative architecture diagram, duplicated examples, and inline time-sensitive version details, and its workflows lack embedded validation checkpoints. Restructuring for tightness and explicit verify steps would lift the two mid-range dimensions.

Suggestions

Trim the 25-line ASCII architecture diagram to a one-line component list and remove the duplicated '/loop 30m' example by cross-referencing the single instance from the Setup Guide.

Move the pinned package version and registry verification date out of the setup steps into a clearly labeled 'verified versions' or 'old patterns' section so time-sensitive details do not penalize the main flow.

Add explicit validation checkpoints to the Setup Guide workflow (e.g., 'verify the MCP memory server responds before enabling crons', 'confirm the first scheduled run completed') instead of deferring verification advice to the Constraints section.

DimensionReasoningScore

Conciseness

Mostly efficient with well-chosen tables and commands, but it carries a ~25-line ASCII architecture diagram, repeats the same '/loop 30m' example twice (Core Components and Setup Step 2), and embeds time-sensitive version/date details ('version 2026.8.31 was verified on the public npm registry on 2026-09-07') outside any deprecated section.

3 / 5

Actionability

Provides mostly executable guidance: concrete '/loop 30m ...', 'cd ... && claude -p ...' commands and a copy-paste-ready pinned MCP JSON config, though some sections (capability lists, 'Replacing Hermes' table) remain descriptive rather than instructive.

4 / 5

Workflow Clarity

The 4-step Setup Guide gives a clear sequence, but validation is advisory and scattered ('Always verify that scheduled tasks completed successfully' lives in Constraints) rather than explicit checkpoints or feedback loops inside the workflow itself.

3 / 5

Progressive Disclosure

Well-organized single-file skill with clear section headers, no nested or buried references, and navigation-friendly structure; at ~265 lines content such as the three example workflows could arguably be split into separate reference files, keeping it below the top anchor.

4 / 5

Total

14

/

20

Passed

Description

83%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 is strong: third-person voice, explicit 'Use when...' trigger clause, concrete capability list, and distinct positioning against named alternatives. It sits just below the top anchors on specificity and trigger coverage because capabilities are stated as feature nouns and a few natural user phrasings are absent.

DimensionReasoningScore

Specificity

Lists several specific capabilities ('persistent memory, scheduled operations, computer use, and task queuing', 'native crons, dispatch, MCP tools') but they are feature nouns rather than concrete actions, leaving minor gaps versus the comprehensive action-level coverage of the top anchor.

4 / 5

Completeness

Clearly answers both what ('Transform Claude Code into a fully autonomous agent system with persistent memory, scheduled operations, computer use, and task queuing') and when ('Use when the user wants continuous autonomous operation, scheduled tasks, or a self-directing agent loop') with explicit trigger phrases.

5 / 5

Trigger Term Quality

Good natural keyword coverage including 'continuous autonomous operation', 'scheduled tasks', 'self-directing agent loop', and framework names 'Hermes' and 'AutoGPT', but misses common variations users might say such as 'automation' or 'runs on its own'.

4 / 5

Distinctiveness Conflict Risk

Naming Hermes/AutoGPT replacements and Claude Code's native crons/dispatch/MCP carve out a distinct niche with clear triggers; only minor overlap risk with generic scheduling or loop skills.

4 / 5

Total

17

/

20

Passed

Validation

93%

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

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository

affaan-m/ECC

Path

skills/autonomous-agent-harness/SKILL.md

Reviewed

Table of Contents

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.