Install
$ agentstack add skill-ariaxhan-the-agent-library-debate-decisions ✓ 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.
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
Debate Protocols
Use debate as a small decision aid, not a format museum. If grep, tests, logs, source files, or official records can decide the question, use them as the judge and skip debate.
Core Rules
- State the resolution in one sentence.
- Assign sides before arguing. For peer agents, the proposer argues against their own proposal and the reviewer argues for it; otherwise use a coin flip or a stable hash of the resolution.
- Treat assigned side as a role, not belief. Concession is allowed and counts as progress.
- The proposer must argue against their own proposal at least once.
- End with a decision record: winner, strongest argument, unresolved risk, evidence that would flip the result, and preserved dissent.
- Make every dissent testable. An objection must name the evidence that would prove or disprove it, or be labeled speculation.
- Cap the loop: one rebuttal cycle by default. If positions do not change after two exchanges, stop and record the unresolved dissent.
Compressed Rounds
Use compressed rounds by default inside normal work:
Resolution:
Pro constructive:
Con constructive:
Crossfire:
Rebuttals:
Weighing:
Decision:
Preserved dissent:
For multi-agent HCOM work, send each role its assignment and require a short constructive plus one attack on its own side.
Read references/termination-and-concession.md when a debate risks looping, when agents disagree after one rebuttal cycle, or when dissent starts becoming performative.
Output Shape
Decision:
Why:
Best opposing argument:
What would flip it:
Next action:
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: ariaxhan
- Source: ariaxhan/the-agent-library
- 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.