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.
npx skills add Suzhou-Tongyuan/sysplorer-mcp-server --skill ty-sysblock-model-debugginggit clone --depth 1 https://github.com/Suzhou-Tongyuan/sysplorer-mcp-serverWrote 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/skills/suzhou-tongyuan/sysplorer-mcp-server/ty-sysblock-model-debugging)<a href="https://agentmods.dev/skills/suzhou-tongyuan/sysplorer-mcp-server/ty-sysblock-model-debugging"><img src="https://agentmods.dev/badge/skills/suzhou-tongyuan/sysplorer-mcp-server/ty-sysblock-model-debugging/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/skills/suzhou-tongyuan/sysplorer-mcp-server/ty-sysblock-model-debugging"><img src="https://agentmods.dev/badge/skills/suzhou-tongyuan/sysplorer-mcp-server/ty-sysblock-model-debugging.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 | $0.00079 | $0.01803 |
| Opus 5.5 | $0.00032 | $0.00721 |
| Sonnet 5.5 | $0.00016 | $0.00361 |
| Haiku 4.5 | $0.00008 | $0.00180 |
Grade A, and why
ty-sysblock-model-debugging 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 19d 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 — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sysblock Model Debugging
Use official structured APIs for topology and parameters in one verified
task-owned copy. Record source/working identities and preserve target settings.
For ordinary Sysblock edits/delivery, reuse ty-sysblock-diagram-modeling API,
layout, save/reload and audit constraints; add signal-skill constraints only for
relevant signal tasks. Only physical-top-level changes need the physical modeling
skill. Loading constraints never reroutes or restarts this debug workflow.
Ordinary construction, editing, layout and tuning remain with their existing skills.
Task Mode and Applicable Constraints
Enter directly from a concrete fault; neither the modeling router nor the debug
router is a prerequisite. Select diagnose_only or repair from the request.
In diagnose-only mode, inspect and, only if authorized, reproduce the needed
stage; report the cause or evidence gap and stop without a patch. Offline review
does not authorize execution. Repairs follow the workflow below.
Record the requested terminal stage and user acceptance before execution. Reuse the existing attributable baseline, valid session, and task-owned working copy; do not reroute, replan, or recreate them when loading another skill's constraints. Construction-stage debugging returns to the original construction task.
Fast Path
For a simple single-model or anchored CCaller fault, keep the workflow near 12 external calls when practical.
- Read the staged model identity, structured state, source files, and hashes once. Do not scan Git, unrelated directories, helper source, or all API docs.
- Extract every user acceptance invariant, including expected values, signs, units, ranges, sample times, stop time, and KPIs.
- Confirm the target through Sysplorer MCP before first live access: use
session_manager(action="probe")when available, thenhealth. Reuse valid same-session evidence; recheck after a switch, disconnection or infrastructure symptom. Offline artifact inspection needs no session call. Resolve unknown API contracts before calling; do not enumerate all APIs. - Read only the failed block, direct neighbors, relevant ports/connections, parameters, and target configuration.
- Form one falsifiable hypothesis. Apply one minimal structured model change or one user-source patch, then read back the changed value/source and hash.
- Rerun the affected stage and final functional acceptance. Build or simulation success alone is insufficient when the KPI is wrong.
What ships with it
12 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/current-mcp-recipes.md 6.9 KB
- references/external-c-debugging.md 5.8 KB
- references/help-doc-index.md 3.1 KB
- references/report-contract.md 2.5 KB
- references/sysblock-diagnostics.md 4.4 KB
- scripts/debug_workspace.py 14 KB runs code
- scripts/extract_debug_evidence.py 11 KB runs code
- scripts/query_sysplorer_help.py 11 KB runs code
- skill_info.json 796 B
- SKILL_ZH.md 5.6 KB
- 使用说明.md 1.5 KB
- 版本说明.md 1.2 KB
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.
- 19d ago First seen · 159 lines · 79 tokens per session scan A 3456d8e011b7
ty-sysblock-model-debugging is a skill published in the GitHub repository Suzhou-Tongyuan/sysplorer-mcp-server (10 stars, last pushed 20d ago), licensed MIT. It adds 79 tokens to every session and 1,803 once invoked, about $0.0003 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 2026-10-13.
Other skills, from other repositories
systematic-debugging
Use when debugging a failing test, build error, or runtime issue that isn't immediately obvious. Guides a 4-phase root cause analysis instead of random fix attempts.
log-error-digest
Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…
cwv-optimizer
Diagnose and fix Core Web Vitals issues on AEM Edge Delivery Services pages. Goes deeper than generic CWV advice by understanding EDS-specific performance patterns including the 100KB LCP budget, E-L-D loading phases, block rendering behavior, and third-party script impact. Produces specific fixes for LCP, CLS, and…
ci-fixer
CI failures - read error, minimal fix, verify.
incident-response-incident-response
Coordinate production incident triage, mitigation, investigation, recovery, and follow-up. Use when responding to outages or other incidents that require structured operational coordination.
diagnosing-missing-recordings
Diagnoses why a session recording is missing or was not captured. Use when a user asks why a session has no replay, why recordings aren't appearing, or wants to troubleshoot session replay capture issues for a specific session ID or across their project. Covers SDK diagnostic signals, project settings, sampling…