agentsop-llm-tool-idempotency
Decision protocol for making side-effectful agent tools idempotent — so when an LLM tool call is retried (timeout, framework resume, user re-run, model…
Sorting a stream of records into a picture
What it does
> One-liner: The LM is at-least-once; the tool must be at-most-once. > Every framework that promises "durable execution" still re-runs node bodies > on resume. Every HTTP client retries on timeout. Every model occasionally > emits the same tool_call twice. Idempotency belongs in the tool, not in a > wish.
Activate this skill when any of the following triggers fire
You don’t need to name it. Ask for the job in your own words — your AI loads the skill when the request matches.
What it can touch
No script files in its folder
Its folder holds text, not programs. It changes how your AI works; it does not bring code of its own.
How we know
Listed the 5 files in the skill’s folder on GitHub; none is a script or program.
Mentions 6 outside web addresses
The instructions point your AI at these sites or APIs. Anything it sends there leaves your computer.
How we know
Web addresses in the SKILL.md, excluding GitHub and documentation sites: docs.aws.amazon.com, docs.crewai.com, docs.stripe.com, docs.temporal.io, microservices.io….
No known risk pattern found
Launch7’s automated check read all 5 text files in this skill and matched none of its risk patterns. This is not a guarantee: a check like this can miss new or disguised tricks, and the author can change the files after the check.
How we know
Read 2026-10-04, commit 6ea799f by pipeline/trust/analyze.py — 40+ patterns for hidden text, prompt injection, downloads that run, credential access and more. It reads; it never runs the skill.
Recently maintained
Its repository has had changes recently.
How we know
Last push to agentsope/SkillAlchemy: 2026-09-02 (32 days before the scan).
Read automatically from the skill’s own files on GitHub on 4 Oct 2026. Launch7 does not run or manually review skills: this lists what the check saw, and nothing here means “safe”. How we check
Add it to your AI
Supported. Upload it once in Claude’s settings. Works on every plan, including Free.
Built in your browser from the SKILL.md on GitHub — nothing passes through Launch7.
In Claude, open Customize → Skills, press +, then Create skill → Upload a skill, and choose the .zip.
Make sure code execution is switched on (Settings → Capabilities). Skills need it.
Start a new chat and ask for the job in your own words — Claude picks the skill up when it fits.
How Claude does this: support.claude.com · checked 28 Sep 2026 · “Supported” means documented by the vendor; Launch7 has not run it end to end
More from agentsope
All from this publisher →- agentsop-agent-topology-selection
Cross-framework enhancement overlay for choosing a multi-agent topology BEFORE writing any agent.
- agentsop-aider
SOP for terminal-based, git-native AI pair programming with Aider (git work-tree + tree-sitter repo-map + edit-format + human-in-loop REPL).
- agentsop-bio-fraud-forensics
Screens biomedical / life-science papers for signs of data fabrication, image manipulation, and statistical anomalies, using the detection techniques…
- agentsop-bounded-loop
Universal discipline for any LM-driven loop — agent retries, plan-act-observe, multi-agent handoffs, optimiser passes, test-fix cycles.
- agentsop-context-scope-discipline
Coder-agent working-file budget discipline: keep the editable working set (files you /add into writable context) under ~25k tokens, separate "read" from…
