skill-gb3h-agent-skills-agent-self-recovery · by Install
$ agentstack add skill-gb3h-agent-skills-agent-self-recovery ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-gb3h-agent-skills-agent-self-recovery) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Agent Self-Recovery
Agents get stuck. The key is recognizing it early and breaking out deliberately.
Stuck Patterns to Watch For
- Retry loop — Same command, same error, >2 attempts
- Fix spiral — Each fix creates a new error, no net progress
- Scope creep drift — Started task A, now working on task G
- Overthinking freeze — Generating analysis without taking action
- Tool misuse — Using the wrong tool repeatedly because the right one isn't obvious
Recovery Protocol
1. Name the Pattern
Stop and explicitly state: "I'm stuck in a [pattern name]." Writing it down breaks the loop.
2. Capture State
Write down:
- What I was trying to do
- What I've tried so far
- What keeps failing and why
3. Change Strategy
Pick ONE:
- Step back: Re-read the original task. Am I solving the right problem?
- Simplify: What's the minimum version of this that could work?
- Ask: Tell the user what's happening and ask for guidance
- Defer: Write a handoff document (see
task-handoff) and stop - Research: Stop guessing, read docs/source code
4. Set a Limit
"I will try this new approach for N steps. If it doesn't work, I'll [escalate/defer/ask]."
Prevention
- After 2 failed attempts at the same thing, pause and assess
- Keep a mental (or written) count of retries
- If you're writing the same command again, something is wrong
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: gb3h
- Source: gb3h/agent-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.