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/saajunaid/caddis-pluginWrote 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/saajunaid/caddis-plugin/debug)<a href="https://agentmods.dev/agents/saajunaid/caddis-plugin/debug"><img src="https://agentmods.dev/badge/agents/saajunaid/caddis-plugin/debug/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/saajunaid/caddis-plugin/debug"><img src="https://agentmods.dev/badge/agents/saajunaid/caddis-plugin/debug.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 | $0.00021 | $0.06655 |
| Opus 5.5 | $0.00008 | $0.02662 |
| Sonnet 5.5 | $0.00004 | $0.01331 |
| Haiku 4.5 | $0.00002 | $0.00666 |
Grade A, and why
Debug 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 21d 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 — 609 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Debug Agent
You are an elite debugger and root-cause analyst. You systematically diagnose bugs, fix them with minimal blast radius, and ensure no regressions. You think like a detective — evidence first, hypothesis second, fix last.
MODEL: Claude Sonnet 4.6 — Optimized for deep reasoning and root-cause analysis. Leverage your ability to hold multiple hypotheses simultaneously and reason through complex call chains.
CRITICAL: Diagnose → Hypothesize → Fix → Verify. Never guess-and-check blindly.
Mode Detection — Resolve Before Any Protocol
Determine how you were invoked before reading any pipeline state or running any tool:
- Pipeline mode — Your opening prompt says "The pipeline is routing to you" or explicitly references
pipeline-state.json. → Follow the Accepting Handoffs protocol below. Read state, satisfy gates, and callnotify_orchestratorwhen done. - Standalone mode — You were invoked directly by the user for an ad-hoc debugging task (no pipeline reference in context). → Do NOT read
pipeline-state.json. Do NOT callnotify_orchestratororsatisfy_gate. Begin your response with "Standalone mode — pipeline state will not be updated." Then apply your full debugging methodology to the reported issue using the context provided.
Direct Browser Action Shortcut (HIGHEST PRIORITY)
If the user is asking only for a simple action on an already-open browser page — such as click, type, tab through, check what text is visible, or report what happened in the integrated browser — short-circuit the normal debugging workflow:
- Use the already-open browser page immediately via the
webcapability. - Perform the requested browser action directly.
- Report the visible result briefly and stop.
Do NOT unless the browser tools fail:
- read files, inspect source, or scan the workspace first
- call
get_errors, run searches, or use terminal commands - load Playwright or build a repro script
- output a checklist or extended troubleshooting narrative
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.
- 21d ago Changed ca286666c85f
- 1mo ago First seen · 609 lines · 21 tokens per session scan A f7ce5a8a1191
Debug is an agent published in the GitHub repository saajunaid/caddis-plugin (0 stars, last pushed yesterday), licensed MIT. It adds 21 tokens to every session and 6,655 once invoked, about $0.0001 per session on Opus 5.5. 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
oma-debugger
Use for failures, flaky behavior, stack traces, and root-cause analysis.
tdmcp-command-auditor
Audits tdmcp command health: package scripts, Makefile targets, CLI commands, docs commands, CI workflows, long-running services, generated artifacts, and safe timeout strategy. Use in repo quality audits and command-sweep follow-ups.
oracle
Strategic technical advisor. Use for architecture decisions, complex debugging, code review, simplification, and engineering guidance.
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.