goose-session-report - GitLite

KO | EN

gitlite — search

#monitoring #electron #openai #react-native #framework #computer-vision #list #web #php #python

goose-session-report

(★ 18)

Turn Goose session logs into a clean, readable summary report.

File Explorer

ZIP Download

- sample-session.jsonl
- - __init__.py
  - cli.py
  - parse.py
  - report.py
- test_report.py
LICENSE
pyproject.toml
README.md

# Use via CDN

jsDelivr

jsDelivr serves any public GitHub repository as a CDN with zero setup. Pick a version and a file to get a ready-to-paste link and snippet.

Version

File

Link

Example

Command Glossary

Commands referenced in this DOCs, explained below.

🔍

`pip install`

View Details ▼

Install Python packages.

pip install {{package1 package2 ...}}

Install one or more packages:

pip install {{package1 package2 ...}} {{[-U|--upgrade]}}

Upgrade all specified packages to the latest version, installing any that are not already present:

pip install {{package}}=={{version}}

Install a specific version of a package:

🔍

`pytest`

View Details ▼

Run Python tests.

pytest {{path/to/test_file1.py path/to/test_file2.py ...}}

Run tests from specific files:

pytest -k {{expression}}

Run tests with names matching a specific [k]eyword expression:

pytest {{[-x|--exitfirst]}}

Exit as soon as a test fails or encounters an error:

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

**Turn Goose session logs into a clean, readable summary report.**

[Goose](https://github.com/block/goose) records each session as a line-delimited
JSON log. Those files are great for the machine but hard for a human to skim.
`goose-session-report` reads a session and produces a tidy **Markdown** or **HTML**
summary: how many messages, who sent them, which tools were used, how often, and
how many tool calls failed.

Handy for reviewing what an agent actually did, sharing a run with a teammate, or
attaching evidence to a report.

Part of the AAIF (Agentic AI Innovation Foundation) open-source toolset for the
agentic AI ecosystem.

## Highlights

- Summarizes messages by role and tool usage with failure counts.
- Tolerant parser: skips unreadable lines and copes with field-name differences
  across Goose versions instead of crashing.
- Markdown or self-contained HTML output.
- Standard library only. No pip dependencies.

## Install

```bash
pip install -e .
```

Requires Python 3.9 or newer.

## Usage

```bash
goose-session-report examples/sample-session.jsonl
```

Write an HTML report:

```bash
goose-session-report examples/sample-session.jsonl --format html -o report.html
```

Goose session files usually live under your Goose data directory (for example
`~/.local/share/goose/sessions/` on Linux). Point the tool at any `.jsonl`
session file.

## Example output

```markdown
# Goose session report

**Messages:** 8
**Tool calls:** 3

## Messages by role

| Role | Count |
|------|-------|
| assistant | 3 |
| tool | 3 |
| user | 2 |

## Tools used

| Tool | Calls | Failures |
|------|-------|----------|
| `shell` | 2 | 1 |
| `text_editor` | 1 | 0 |
```

## Notes on formats

Goose has evolved its log schema over time. This tool normalizes several known
shapes for message content and tool calls (`toolRequest`/`tool_request`/`toolCall`
and their response counterparts). If your session uses a field this parser does
not recognize, it will skip that detail rather than fail. Contributions to widen
coverage are welcome.

## Development

```bash
pip install -e .
pytest
```

## License

MIT. See [LICENSE](LICENSE).