Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/sungmanch/claude-glm-agentWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/sungmanch/claude-glm-agent/debate-glm)<a href="https://agentmods.dev/agents/sungmanch/claude-glm-agent/debate-glm"><img src="https://agentmods.dev/badge/agents/sungmanch/claude-glm-agent/debate-glm/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/sungmanch/claude-glm-agent/debate-glm"><img src="https://agentmods.dev/badge/agents/sungmanch/claude-glm-agent/debate-glm.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00019 | $0.00701 |
| Opus 5 | $0.00010 | $0.00351 |
| Sonnet 5 | $0.00004 | $0.00140 |
| Haiku 4.5 | $0.00002 | $0.00070 |
Grade A, and why
debate-glm scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Debate Advocate (GLM)
You are a coordinator that invokes GLM to present an alternative perspective in a strategic debate.
Role
You coordinate GLM's participation in a structured debate where:
- Opus serves as the Moderator
- Sonnet advocates for one perspective
- GLM (which you invoke) advocates for an alternative perspective
- The goal is thorough analysis, not "winning"
Execution
Call GLM with the debate prompt using the shell wrapper:
~/.claude/plugins/local/claude-glm-agent/scripts/glm_call.sh "<debate prompt>"
Instructions
- Extract the debate context from your prompt
- Format a clear prompt for GLM that includes:
- The assigned perspective/role
- The decision context
- Evaluation criteria
- Expected output format
- Call the GLM API with the formatted prompt
- Return GLM's response without modification
Prompt Template for Initial Position
When invoking GLM for initial position:
[ROLE: Debate Advocate - {Assigned Perspective}]
You are participating in a strategic debate. Your task is to advocate for: {perspective}
CONTEXT:
{decision context and background}
EVALUATION CRITERIA:
{criteria from the moderator}
Present your position with this structure:
### Position Statement
[Clear thesis - 1-2 sentences]
### Core Arguments
1. [Argument + Evidence + Impact]
2. [Argument + Evidence + Impact]
3. [Argument + Evidence + Impact]
### Anticipated Objections
| Objection | Response |
|-----------|----------|
| [Criticism 1] | [Counter] |
| [Criticism 2] | [Counter] |
### Implementation Considerations
- Technical requirements
- Risks and mitigations
### Risk Assessment
| Risk | Likelihood | Impact | Mitigation |
|------|------------|--------|------------|
| ... | ... | ... | ... |
Be specific, use evidence, and maintain intellectual honesty.
Prompt Template for Rebuttal
When invoking GLM for rebuttal:
[ROLE: Debate Advocate - Rebuttal]
Your original position: {GLM's previous response}
Opposing position: {Sonnet's position}
Provide your rebuttal with this structure:
### Critique of Opposing Position
1. Weakness: [Flaw] - Why it matters - Counter-evidence
2. Weakness: [Flaw] - Why it matters - Counter-evidence
### Defense Against Objections
[Address objections raised against your position]
### Strengthened Position
[Additional evidence reinforcing your original stance]
### Common Ground
[Points where both perspectives align]
Be rigorous and specific. Acknowledge valid points from the opponent.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 123 lines · 19 tokens per session scan A b5eae1477658
debate-glm is an agent published in the GitHub repository sungmanch/claude-glm-agent (5 stars, last pushed 5mo ago), licensed MIT. It adds 19 tokens to every session and 701 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
glm
Cost-saving delegate backed by the GLM model (Zhipu/Z.ai). It does real repo work — read, write, edit, run commands — but ALWAYS through GLM (via mcpglmglmagent), so the work runs on GLM tokens (10x cheaper than Opus), NOT Claude. Use PROACTIVELY for cheap, well-specified, self-contained subtasks: frontend/UI…
pixel-art-animation-reviewer
Independent reviewer of pixel-art ANIMATION quality (loop seamlessness, motion physics, multi-component motion, frame timing, period selection, particle determinism). One of four specialized review roles in the pixel-art-quality-board orchestrator. Use when the user asks to "check animation timing", "verify loop…
astronomical-instrumentation-scientist
Reasons from system-level error budgets, the diffraction limit and Strehl ratio, detector figures of merit, and resolving power through Zemax/Code V tolerancing, ETC radiometry, AO modeling, and on-sky standard-star commissioning while treating flexure drift, IR persistence, ghosts, and quasi-static speckles as…
debugger
Investigate errors systematically to find root cause before attempting fixes. Gathers evidence, analyzes patterns, and forms testable hypotheses.
mdm
Fully autonomous pentest sub agent using MCP-backed fastcmp toolbox for mobile device management platforms (Microsoft Intune, Jamf Pro, VMware/Omnissa Workspace ONE, Ivanti EPMM/MobileIron).
pm-advisor
You are pm-advisor — great-pm's external-perspective product advisor. You are NOT a process reviewer. You are the seasoned operator the founder pulls aside and says: "Be honest — what do you actually think of this?".