Logfire

Logfire Log Query Skill

Query Pydantic Logfire for logs, traces, errors, and performance data.

Setup

Requires LOGFIRE_READ_TOKEN environment variable (from .env or exported).

Bundled Scripts

Run scripts from the skill's scripts directory using uv run:

Replace {SKILL_DIR} with the actual skill directory path shown in the "Base directory" header.

Script Options

errors.py

--hours N      Hours to look back (default: 24)
--limit N      Max results (default: 20)
--file PATH    Filter by filepath in stacktrace

slow.py

--hours N      Hours to look back (default: 24)
--min-ms N     Minimum duration in ms (default: 1000)
--limit N      Max results (default: 20)
--endpoint X   Filter by endpoint name

recent.py

--minutes N    Minutes to look back (default: 5)
--limit N      Max results (default: 30)

search.py

--hours N      Hours to look back (default: 24)
--limit N      Max results (default: 20)
--span         Search span_name instead of message
--verbose      Show full messages

query.py

--trace ID     Look up specific trace
--span ID      Filter by span ID (requires --trace)
--age N        Minutes to look back (default: 1440)
--json         Output raw JSON

endpoints.py

--hours N      Hours to look back (default: 24)
--limit N      Max endpoints (default: 20)
--errors       Only show endpoints with errors

Records Table Schema

Key columns in the records table:

• start_timestamp - When the span started
• span_name - Name of the span (e.g., endpoint name)
• message - Log message
• duration - Span duration in seconds (multiply by 1000 for ms)
• is_exception - Boolean, true if this span has an exception
• trace_id - Trace identifier (use for grouping related spans)
• span_id - Individual span identifier
• attributes - JSON object with span attributes

Example Usage

User: /logfire errors → Run: uv run python {SKILL_DIR}/scripts/errors.py

User: /logfire trace 019c22c6b9fd5b710ca67ed52055d835 → Run: uv run python {SKILL_DIR}/scripts/query.py --trace 019c22c6b9fd5b710ca67ed52055d835

User: /logfire search "upload" --span → Run: uv run python {SKILL_DIR}/scripts/search.py "upload" --span

User: /logfire slow --min-ms 2000 → Run: uv run python {SKILL_DIR}/scripts/slow.py --min-ms 2000

Converted and distributed by TomeVault — claim your Tome and manage your conversions.