Claude Code Plugins

Community-maintained marketplace

Feedback

logfire-read

@michaeltschreiber/logger
4
0

Query Logfire logs via the Logfire query API using read tokens and JSONL output. Use when you need to fetch or filter Logfire records, build log queries, or provide agent-ready log retrieval from this repo. Triggers: Logfire read, query API, JSONL logs, read token, log retrieval, log queries.

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name logfire-read
description Query Logfire logs via the Logfire query API using read tokens and JSONL output. Use when you need to fetch or filter Logfire records, build log queries, or provide agent-ready log retrieval from this repo. Triggers: Logfire read, query API, JSONL logs, read token, log retrieval, log queries.

Logfire Read Skill

Use this skill to read Logfire records with JSONL output using the repo’s reader utilities.

Quick Start

  1. Load .env from the repo root:
set -a
. ./.env
set +a
  1. Run the JSONL reader:
python scripts/logfire-fetch.py

Or use the bundled script:

./.codex/skills/logfire-read/scripts/read_logs.sh --message-like "Processing failed" --limit 10

Agent-Friendly CLI

Use the agent helper for targeted queries:

python scripts/logfire-read-agent.py --since 2026-01-11T00:00:00Z --message-like "Processing failed" --limit 10

Module API

Use query_logfire() for programmatic access:

from logger import query_logfire

rows = query_logfire(jsonl=True, level=21, message_like="Processing failed", limit=10)
for row in rows:
    print(row)

Environment Variables

Required:

  • LOGFIRE_READ_URL (e.g., https://logfire-us.pydantic.dev/v1/query)
  • LOGFIRE_READ_TOKEN

Optional filters:

  • LOGFIRE_READ_COLUMNS
  • LOGFIRE_READ_LIMIT
  • LOGFIRE_READ_SINCE
  • LOGFIRE_READ_LEVEL
  • LOGFIRE_READ_MESSAGE_LIKE
  • LOGFIRE_READ_TRACE_ID
  • LOGFIRE_READ_SPAN_ID
  • LOGFIRE_READ_JSONL=true

Output Format

  • JSONL output returns one JSON object per record.
  • level is converted to severity strings (trace, debug, info, warn, error, fatal).