Borrowing it
Nothing to install: this file belongs to gradusnikov/eclipse-chatgpt-plugin. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/gradusnikov/eclipse-chatgpt-plugin/main/.claude/skills/eclipse-debug/SKILL.mdgit clone --depth 1 https://github.com/gradusnikov/eclipse-chatgpt-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/skills/gradusnikov/eclipse-chatgpt-plugin/eclipse-debug)<a href="https://agentmods.dev/skills/gradusnikov/eclipse-chatgpt-plugin/eclipse-debug"><img src="https://agentmods.dev/badge/skills/gradusnikov/eclipse-chatgpt-plugin/eclipse-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/skills/gradusnikov/eclipse-chatgpt-plugin/eclipse-debug"><img src="https://agentmods.dev/badge/skills/gradusnikov/eclipse-chatgpt-plugin/eclipse-debug.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 21 Skill allows unbounded resource consumption (API calls, storage, compute). Without rate limits or quotas, a compromised or misbehaving agent can cause denial-of-service or cost overruns.Fix: Set explicit rate limits, timeouts, and resource quotas for API calls, file operations, and compute. Implement circuit breakers for runaway loops.
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 | $0.00037 | $0.01201 |
| Opus 5.5 | $0.00015 | $0.00480 |
| Sonnet 5.5 | $0.00007 | $0.00240 |
| Haiku 4.5 | $0.00004 | $0.00120 |
Grade A, and why
eclipse-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 2mo 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 — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Debug Java Applications in Eclipse
Full interactive debugging using Eclipse's JDT debugger. Set breakpoints, step through code, inspect variables, evaluate expressions, and hot-swap changes.
Breakpoints
- toggleBreakpoint — Set or remove a line breakpoint. Provide
projectName,typeName(fully qualified), andlineNumber. - setConditionalBreakpoint — Set a breakpoint with a condition. Only triggers when the condition is true. Also supports
hitCount(trigger after N hits). - listBreakpoints — Show all breakpoints with location, enabled status, and conditions.
- removeAllBreakpoints — Clear all breakpoints.
Launch
- debugJavaApplication — Launch in debug mode. Same parameters as
runJavaApplicationbut defaults totimeout=0(background). The app stops at breakpoints. - launchConfiguration — Debug an existing saved launch configuration by name with
mode="debug", preserving its classpath, VM args, environment variables, working directory, and agent settings (e.g. JRebel). UselistLaunchConfigurationsto find the name. Breakpoints set viatoggleBreakpointstill apply.
Stepping (requires suspended thread)
- stepOver — Execute current line, don't enter method calls.
- stepInto — Enter the method call on the current line.
- stepReturn — Run until the current method returns.
- resumeDebug — Continue execution until next breakpoint or termination.
Inspection (requires suspended thread)
- getStackTrace — Show all threads with call stack. For the top frame, also shows local variables with their values and types.
- evaluateExpression — Evaluate any Java expression in the context of the suspended frame. Examples:
myList.size(),x + y,this.toString(),System.getProperty("user.dir").
Hot Code Replace
- hotCodeReplace — Push code changes into the running debug session without restarting. Triggers an incremental build and the JVM reloads changed classes.
Structural Reloads
- restartMcpServers — Rebuild the HTTP MCP server registry after a 500–30,000 ms delay. The delay lets the invoking response complete; clients may need to reconnect.
- reloadWorkspaceBundle — Schedule an OSGi update for a bundle whose symbolic name maps to an open workspace project. System bundles, fragments, and installed-only bundles are rejected.
- Use
hotCodeReplacefor supported method-body changes. UsereloadWorkspaceBundlewhen structural changes (new methods, fields, or MCP tools) cannot be hot-swapped, thenrestartMcpServersif the HTTP registry still needs rebuilding.
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.
- 2mo ago First seen · 94 lines · 37 tokens per session scan A d6de924820d7
eclipse-debug is a skill published in the GitHub repository gradusnikov/eclipse-chatgpt-plugin (175 stars, last pushed 9d ago), licensed MIT. It adds 37 tokens to every session and 1,201 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 2026-08-30.
Other skills, from other repositories
mcp-java-dev-tools-bug-drill
Discover bounded Java method drill candidates or run a Probe-guided diagnostic drill for one selected method and return an evidence-backed Markdown report.
mcp-java-dev-tools-bug-fix
Create a proposal-only, evidence-backed Java bug-fix diagnosis from issue context and behavioral reproduction steps. Use when expected behavior and how to reproduce are known, even when endpoint, class, method, logs, or technical request details are not yet known.
mcp-java-dev-tools-line-probe-run
Run strict single-line JVM probe verification. Use for line-hit checks, route validation, and fail-closed probe execution.
mcp-java-dev-tools-performance-suite
Run MCP-first Java performance suites with required strict line-hit verification, concurrency-based workload contracts, and deterministic threshold evaluation. Use when executing or reviewing performance-suite Artifacts under .mcpjvm/.../plans/performance or execution profiles with suiteType=performance.
mcp-java-dev-tools-failure-lens
Investigate a pasted Java stack trace against an attached Sidecar Agent using bounded guided or hands-off reproduction. Use when a user wants to reproduce a Java exception with runtime Probe evidence; never use for static-only diagnosis, CI/regression execution, generic replay, or automatic Bug Fix handoff.
mcp-java-dev-tools-performance-plan-crafter
Create or update deterministic performance plans (metadata.json, contract.json, plan.md) under .mcpjvm/.../plans/performance. Use when authoring workload-centric Java performance-suite Artifacts with required strict line-hit verification.