debug-proxy

A troubleshooting guide for Gantry's MCP proxy, the layer that passes tool requests between an agent and the server. It covers logs, request history, sessions, caches, safety checks, retries, and circuit breakers.

In plain words
What is it for?
Debugging failed or delayed tool calls, stale cached data, false safety warnings, retry storms, circuit-breaker trips, and proxy instability.
Why use it?
It provides a structured way to investigate unexpected tool results or stuck agents before changing code, and helps identify shared state or server-health problems.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/geleynse/gantry/debug-proxy
Any agent
npx skills add geleynse/gantry --skill debug-proxy
Clone the repo
git clone --depth 1 https://github.com/geleynse/gantry

Made for: Claude Code, Codex.

Per session 49 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,612 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00049 $0.02612
Opus 5 $0.00024 $0.01306
Sonnet 5 $0.00010 $0.00522
Haiku 4.5 $0.00005 $0.00261

Measured 2d ago against content hash 1b32f27e5bd4, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

debug-proxy 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 2d 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.

.claude/skills/debug-proxy/SKILL.md · 193 lines

How it starts

The opening of the file, as written. The whole thing — 193 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Debugging the Proxy — Gantry

Systematic approach for "an agent got a weird result" or "an agent is stuck" reports. Work top-down: logs → tool-call history → session state → server health, before touching guard/injection code.

1. Turn up logging

LOG_LEVEL=debug bun run dev      # or set in the environment before `bun dist/index.js`

Format (from src/lib/logger.ts, createLogger(category)):

[2026-07-01T12:34:56.789Z] [DEBUG] [passthrough] jump BEFORE | agent: rust-vane, system: sol_prime, dest: void_reach, fuel: 42, tick: 118, server_tick: 119, drift: 1

[LEVEL] [category] message | key: value, key2: value2. Useful categories to grep for: pipeline, passthrough, registry, gantry-v2, compound-tools, fuel-floor-guard, circuit-breaker, session, instability-hints, instability-metrics, proxy (constants/reformat). console.error is used for WARN/ERROR, console.log for INFO/DEBUG — pipe accordingly if you're filtering stdout vs stderr.

2. Read the tool-call history — two-phase pending→complete records

Every tool call writes a row to the proxy_tool_calls SQLite table ($FLEET_DIR/data/fleet.db) via tool-call-logger.ts:

  • logToolCallStart(agent, toolName, args, opts) inserts a row with status = "pending" and returns a numeric pendingId.
  • logToolCallComplete(pendingId, agent, toolName, result, durationMs, opts) updates that same row to status = "complete" or "error", filling in result_summary, duration_ms, success, error_code.

Columns worth querying directly: agent, tool_name, args_summary, result_summary, success, error_code, duration_ms, is_compound, status, trace_id, parent_id (compound tools log sub-calls with parent_id pointing at the parent row — same mechanism PrayerLang uses to build its execution tree), timestamp.

A row stuck at status = "pending" with no matching complete update means the call never returned — either it's still in flight, or the process crashed/restarted mid-call (the pendingId insert survives; nothing ever calls logToolCallComplete for it). That's a strong signal to check for a server restart or an unhandled exception around that timestamp.

Read the full file on GitHub · 193 lines

Changes

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.

  1. 2d ago First seen · 193 lines · 49 tokens per session scan A 1b32f27e5bd4

Subscribe to this mod's changes

debug-proxy is a skill published in the GitHub repository geleynse/gantry (3 stars, last pushed 14d ago), licensed MIT. It adds 49 tokens to every session and 2,612 once invoked, about $0.0002 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens