Skip to main content

ATP plugin: run agent-eval-case methodology cases through the platform

Project description

atp-method

ATP plugin that runs method/ agent-eval-case methodology cases through the platform: a schema model + a loader that maps each case to an ATP TestDefinition, plus (in later slices) a methodology-aware evaluator and the format-dispatch registration so atp test method/cases/*.yaml just works.

See the design in spec/atp-method-plugin.md.

Status

  • schema model + loader (case → TestDefinition)
  • AgentEvalCaseEvaluator (critical_check then rubric)
  • register() + source dispatch + E2E (atp test method/cases/<case>.yaml)

Usage

Installed as a plugin, the platform runs methodology cases directly — a single case or a whole sweep (directory):

atp test method/cases/req-extraction --adapter=http \
  --adapter-config endpoint=http://agent:8000/execute,allow_internal=true --runs=10

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

atp_method-0.1.0.tar.gz (12.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

atp_method-0.1.0-py3-none-any.whl (10.1 kB view details)

Uploaded Python 3

File details

Details for the file atp_method-0.1.0.tar.gz.

File metadata

  • Download URL: atp_method-0.1.0.tar.gz
  • Upload date:
  • Size: 12.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for atp_method-0.1.0.tar.gz
Algorithm Hash digest
SHA256 41f7954d96d3ca6a1b6e9fa59f15330db358ee39be2450267b703ef10a4a9058
MD5 acad26a816a8f98da67f2b5fb6d7c5d0
BLAKE2b-256 c48d6fa62e2ac38d87955e5459e4c3bd6ab2f5b843fdd580ed6387867c6e5f52

See more details on using hashes here.

Provenance

The following attestation bundles were made for atp_method-0.1.0.tar.gz:

Publisher: atp-method-publish.yml on andrei-shtanakov/atp-platform

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file atp_method-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: atp_method-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 10.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for atp_method-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 67486fc379338684d4b371d89bc1a9da2b8078e6c96d6f75e5ede5a36afbbded
MD5 1f9d2e47c22add014052380694b6d6eb
BLAKE2b-256 a28a7dd3c69089d5c12d2c2b425ce444cc800020ccbdca509140a87c4ba95e0d

See more details on using hashes here.

Provenance

The following attestation bundles were made for atp_method-0.1.0-py3-none-any.whl:

Publisher: atp-method-publish.yml on andrei-shtanakov/atp-platform

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page