error-recovery
Recover Failed Subagent Tasks
Subagent workflows can stall when errors, unclear requirements, or contract issues appear. This skill gives retry, context, split, escalation, and abort patterns.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Review the Skillstore skill "error-recovery" from https://skillstore.io/skills/clouder0-error-recovery.md and its manifest at https://skillstore.io/api/skills/clouder0-error-recovery/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "error-recovery". A subagent timed out and returned no output.
Expected outcome:
Classifies the issue as transient. Recommends a simple retry, tracks the attempt count, and stops after three failed attempts.
Using "error-recovery". An implementer says the requirements are unclear and asks questions in its output.
Expected outcome:
Classifies the issue as a context gap. Recommends adding related files, examples, and clarification before retrying.
Using "error-recovery". A task reports a boundary violation and cannot continue.
Expected outcome:
Classifies the issue as boundary or contract related. Recommends escalation to a contract resolver with the blocked reason and affected contract paths.
Security Audit
SafeAll static findings are false positives caused by Markdown backticks, fenced examples, and ordinary failure-state terms in SKILL.md. The skill is documentation for retry, context enhancement, scope reduction, escalation, and abort patterns, with no command execution, reconnaissance, prompt injection, or data exfiltration intent.
Risk Factors
⚙️ External commands (19)
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/clouder0-error-recovery/audits/8?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/clouder0-error-recovery?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/clouder0-error-recovery?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/clouder0-error-recovery/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/clouder0-error-recovery.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
Clouder0. (2026). error-recovery security audit report (audit version 8) [Author version unspecified]. Skillstore. https://skillstore.io/skills/clouder0-error-recovery/audits/8BibTeX citation
@techreport{clouder0-clouder0-error-recovery-2026,
author = {Clouder0},
title = {error-recovery security audit report (audit version 8)},
institution = {Skillstore},
year = {2026},
number = {8},
url = {https://skillstore.io/skills/clouder0-error-recovery/audits/8},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "error-recovery security audit report (audit version 8)"
version: "unspecified"
type: report
authors:
- name: "Clouder0"
date-released: "2026-07-06"
url: "https://skillstore.io/skills/clouder0-error-recovery/audits/8"
identifiers:
- type: other
value: "skillstore:clouder0-error-recovery:audit:8"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: HighWhat You Can Build
Recover Multi-Agent Workflows
Use structured retry and escalation choices when one subagent fails inside a larger workflow.
Improve Coding Agent Reliability
Give Claude, Codex, or Claude Code a clear response plan for timeout, confusion, and partial work.
Standardize Failure Reports
Create consistent recovery notes when repeated attempts fail and a task needs redesign.
Try These Prompts
Use the error-recovery skill. A subagent timed out and returned no useful output. Classify the failure and recommend the next retry step.
Use the error-recovery skill. The previous subagent said the task was unclear. Recommend what context to add before the next attempt.
Use the error-recovery skill. A broad task produced partial work and then failed. Break the remaining work into smaller subagent tasks.
Use the error-recovery skill. A task is blocked by a boundary or contract issue. Choose the escalation path and define the resolution request.
Best Practices
- Classify the failure before choosing a recovery action.
- Preserve partial work and add exact context for the next attempt.
- Stop after repeated failed recovery and produce a clear report.
Avoid
- Retrying the same broad task without using the failure category.
- Escalating every failure before trying an appropriate recovery step.
- Discarding partial work when scope reduction can reuse it.