SkillRepo
Sign InSign Up
Catalog
affaan-m/continuous-agent-loop

affaan-m

continuous-agent-loop

Patterns for continuous autonomous agent loops with quality gates, evals, and recovery controls.

global
origin:ECC
New~255
v1.3Saved Jul 14, 2026

Continuous Agent Loop

This is the v1.8+ canonical loop skill name. It supersedes autonomous-loops while keeping compatibility for one release.

Loop Selection Flow

Start
  |
  +-- Need strict CI/PR control? -- yes --> continuous-pr
  |
  +-- Need RFC decomposition? -- yes --> rfc-dag
  |
  +-- Need exploratory parallel generation? -- yes --> infinite
  |
  +-- default --> sequential

Combined Pattern

Recommended production stack:

  1. RFC decomposition (ralphinho-rfc-pipeline)
  2. quality gates (plankton-code-quality + /quality-gate)
  3. eval loop (eval-harness)
  4. session persistence (nanoclaw-repl)

Failure Modes

  • loop churn without measurable progress
  • repeated retries with same root cause
  • merge queue stalls
  • cost drift from unbounded escalation

Recovery

  • freeze loop
  • run /harness-audit
  • reduce scope to failing unit
  • replay with explicit acceptance criteria
Files1
1 files · 1.0 KB

Select a file to preview

Overall Score

52/100

Grade

C

Adequate

Safety

80

Quality

45

Clarity

68

Completeness

35

Summary

Continuous Agent Loop teaches patterns for orchestrating autonomous agent execution with quality gates, evaluations, and recovery controls. It provides a flow diagram for selecting loop strategies (sequential, PR-based, RFC-decomposed, exploratory) and recommends a production stack combining RFC pipelines, code quality gates, eval harnesses, and session persistence. The skill identifies failure modes (loop churn, repeated retries, merge stalls, cost drift) and recovery procedures (freezing loops, auditing, scope reduction, replay).

Detected Capabilities

agent coordinationworkflow orchestrationquality gate definitionevaluation harness integrationfailure mode documentationrecovery procedure guidance

Trigger Keywords

Phrases that MCP clients use to match this skill to user intent.

autonomous agent loopcontinuous executionagent quality gatesloop recoveryagent orchestrationeval harnessagent failure modes

Use Cases

  • Orchestrate multi-step agent tasks with failure recovery
  • Select appropriate loop strategy for CI/PR workflows
  • Implement quality gates and evaluation loops in agent pipelines
  • Debug and recover from infinite loops or merge queue stalls
  • Balance exploration vs. deterministic execution in autonomous workflows

Quality Notes

  • Skill is highly conceptual with reference to external tools (ralphinho-rfc-pipeline, plankton-code-quality, eval-harness, nanoclaw-repl) but provides no procedural steps for implementation
  • Loop selection flow is clear and useful for routing decisions
  • Failure modes are well-identified and practical
  • Recovery procedures are documented but lack implementation details (e.g., what does '/harness-audit' actually do?)
  • Missing concrete examples of how to configure quality gates, set acceptance criteria, or implement the recommended stack
  • No guidance on monitoring loop health or defining 'measurable progress' metrics
  • References internal tool names (ralphinho, plankton, nanoclaw) that may not be accessible or explained
  • Skill reads more as a reference architecture than a how-to guide; execution path is unclear
Model: claude-haiku-4-5-20251001Analyzed: Jul 14, 2026

Reviews

Add this skill to your library to leave a review.

No reviews yet

Be the first to share your experience.

Version History

v1.3

Content updated

2026-07-14

Latest
v1.2

Content updated

2026-04-20

v1.1

Content updated

2026-04-12

v1.0

Seeded from github.com/affaan-m/everything-claude-code

2026-03-16

Use affaan-m/continuous-agent-loop in your dev environment

Start a Developer trial

Command Palette

Search for a command to run...

affaan-m/continuous-agent-loop | SkillRepo