agent-loop

Start, monitor, and cancel durable multi-agent coding loops via Inngest. Use when the user wants to run autonomous coding workloads, execute a PRD with multiple stories, kick off an AFK coding session, have agents implement features from a plan, or manage running loops. Triggers on "start a coding loop", "run this PRD", "implement these stories", "go AFK and code this", "check loop status", "cancel the loop", "joelclaw loop", or any request for autonomous multi-story code execution.

Quality

88%

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

Quality

Content

81%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 dense, actionable SKILL.md that excels at workflow clarity and validation gates — appropriate for a batch, potentially-destructive automation skill — with concrete commands and schemas throughout. The main improvement space is trimming provenance/framing prose and splitting the larger reference tables into bundled files.

Suggestions

Move the Event Schema and Source Files tables into reference files under ./references/ and link to them from the body to tighten the main SKILL.md and improve progressive disclosure.

Remove or condense the 'Source receipt' blockquote; provenance framing does not aid execution and competes with operational context.

Replace the 'read codex exec --help' deferral with the actual invocation flags so implementor dispatch is fully copy-paste ready across platforms.

Dimension	Reasoning	Score
Conciseness	Mostly lean and operational — copy-paste commands, schemas, and gotchas with no basic-concept padding — but the 'Source receipt' provenance blockquote and a few framing sentences could be trimmed without losing actionability.	4 / 5
Actionability	Provides concrete, executable guidance throughout (quick-start CLI commands, a complete PRD JSON template, exact deploy/telemetry commands), with minor gaps such as deferring codex flags to '--help' and Mac-only OrbStack instructions.	4 / 5
Workflow Clarity	The PLANNER→IMPLEMENTOR→REVIEWER→JUDGE pipeline is clearly sequenced with an explicit retry/skip feedback loop, a verification preflight checklist, and a numbered deploy flow with a rollout-status validation checkpoint.	5 / 5
Progressive Disclosure	Well-organized into clearly signaled sections with one well-marked external reference (loop-nanny); the inline Event Schema and Source Files tables are reference material that could live in separate files, but they are kept concise and navigable.	4 / 5
	Total	17 / 20 Passed

Description

95%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.

A strong, third-person description that states concrete capabilities and gives rich, natural trigger phrasing covering both the 'what' and the 'when'. The only minor gap is that the action list in the description is slightly narrower than the full pipeline the skill actually drives.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions — 'Start, monitor, and cancel durable multi-agent coding loops' — but stops short of the fuller action set (implement/review/judge) that the body covers, leaving minor coverage gaps.	4 / 5
Completeness	Explicitly answers both 'what' (start/monitor/cancel durable multi-agent coding loops via Inngest) and 'when' (a 'Use when...' clause plus a 'Triggers on...' clause with concrete trigger phrases).	5 / 5
Trigger Term Quality	Comprehensive natural trigger phrases users would actually say — 'start a coding loop', 'run this PRD', 'go AFK and code this', 'check loop status', 'cancel the loop', 'joelclaw loop' — including synonyms and the product name.	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche — durable multi-agent coding loops via Inngest/joelclaw — with distinct product-specific triggers that minimize overlap with generic coding skills.	5 / 5
	Total	19 / 20 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
relative_links	Relative link issues: 1 suspicious	Warning

	Total	15 / 16 Passed

Repository: joelhooks/joelclaw
Path: skills/agent-loop/SKILL.md
Commit: 7094ced

Reviewed: 3 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.