debug

A debugging command that uses Logfire trace data to investigate application errors. A trace is a recorded path showing what happened during an operation, often across several components.

In plain words
What is it for?
Use it to find recent exceptions in a file, inspect related trace records, query Logfire data, and suggest fixes based on the observed failure.
Why use it?
It helps identify the cause and surrounding context of production failures instead of relying only on an error message or source code.

Command

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 commands/pydantic/skills/debug
Clone the repo
git clone --depth 1 https://github.com/pydantic/skills
Per session 13 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 529 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 $0.00013 $0.00529
Opus 5 $0.00006 $0.00264
Sonnet 5 $0.00003 $0.00106
Haiku 4.5 $0.00001 $0.00053

Measured 2d ago against content hash ab07946c35ef, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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 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.

plugins/logfire/commands/debug.md · 46 lines

How it starts

The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/debug

Use the Logfire MCP server to investigate errors and debug issues using real trace data.

If the user only asks to open, view in browser, use the live view, open Explore, or get a Logfire link, use logfire-ui instead of this query-first debugging workflow. For ambiguous prompts such as "show recent errors", ask whether they want UI or query analysis first.

Prerequisites

The Logfire MCP server must be connected (this plugin configures it automatically). The user must have run logfire auth or set LOGFIRE_TOKEN for their project.

Workflow

When the user reports an error in a specific file

  1. Call the find_exceptions_in_file MCP tool with the file path to get the 10 most recent exceptions from that file.
  2. Show the user the exceptions found - include the exception type, message, and when it occurred.
  3. If a specific exception is relevant, use arbitrary_query to dig deeper into the surrounding trace context.
  4. Suggest a fix based on the exception details and trace data.

When the user asks about general issues or wants debugging analysis

  1. Call get_logfire_records_schema to understand the available columns and tables.
  2. Use arbitrary_query to run SQL queries against the OpenTelemetry data. Common queries:
    • Recent errors: SELECT start_timestamp, message, attributes FROM records WHERE level = 'error' ORDER BY start_timestamp DESC LIMIT 20
    • Slow operations: SELECT span_name, duration, start_timestamp FROM records WHERE duration > interval '1 second' ORDER BY duration DESC LIMIT 20
    • Specific endpoint issues: SELECT * FROM records WHERE attributes->>'http.route' = '/api/users' AND level = 'error' ORDER BY start_timestamp DESC LIMIT 10
  3. Present findings and suggest fixes.

When analysis identifies a trace the user wants to share

  1. Use the configured Logfire link tool, such as project_logfire_link(trace_id=trace_id, project=project), to generate a shareable URL for that specific trace.
  2. Share the link with the user.

Read the full file on GitHub · 46 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 · 46 lines · 13 tokens per session scan A ab07946c35ef

Subscribe to this mod's changes

debug is a command published in the GitHub repository pydantic/skills (127 stars, last pushed 2d ago), licensed MIT. It adds 13 tokens to every session and 529 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 2026-08-30.