Add reliable-agent-loops skill#285
Open
vikast908 wants to merge 1 commit into
Open
Conversation
Covers the three failure modes that cause most agent-loop incidents in production: unbounded iteration, swallowed tool errors, and non-idempotent side effects. Adds loop bounding, structured error feedback, idempotent retries, gating of irreversible actions, resumable state, and per-run tracing, each with a Verification checklist. Follows the standard skill anatomy and references security-and-hardening and observability-and-instrumentation instead of duplicating them. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
A new
reliable-agent-loopsskill on building LLM agent loops that hold up in production.Why
The collection already covers debugging, observability, and security, but not the agent loop itself. The three failures behind most agent-loop incidents aren't addressed anywhere in the repo: iteration that never terminates, tool errors that get swallowed, and side effects that run more than once (the classic double-charge after a retry).
What's in it
Each step has BAD/GOOD code, and the skill closes with a Verification checklist.
Conventions followed
name+descriptiononly; the description leads with what it does, thenUse whentriggers.SKILL.md, no supporting files.security-and-hardening(tool authz, injection through tool output) andobservability-and-instrumentation(logging/metrics) rather than duplicating them.Related: I've opened a second PR adding
evaluating-llm-output, the eval counterpart for catching silent regressions. Happy to adjust naming, scope, or split anything.🤖 Generated with Claude Code