Skip to content

ci: exécuter les 199 tests et le lint à chaque push/PR - #3

Merged
kadohack12 merged 1 commit into
mainfrom
ci/github-actions
Oct 31, 2026
Merged

kadohack12 merged 1 commit into
mainfrom
ci/github-actions

Conversation

@kadohack12

Copy link
Copy Markdown
Owner

Objectif

Ce dépôt a 199 tests et un lint ruff configuré, mais aucun pipeline ne les exécutait. Cette PR ajoute une CI qui valide chaque push et chaque PR.

Ce qui est ajouté

.github/workflows/ci.yml : lint (ruff format --check + ruff check) et pytest, sur une matrice Python 3.11 / 3.12 / 3.13 (les versions déclarées dans pyproject.toml).

strategy:
  matrix:
    python-version: ['3.11', '3.12', '3.13']
steps:
  - pip install -e ".[dev]"
  - ruff format --check . && ruff check .
  - python -m pytest

Vérifié localement (Python 3.13)

  • pip install -e ".[dev]" : OK
  • python -m pytest : 199 passed
  • ruff check . : All checks passed!
  • ruff format --check . : conforme

Cette PR a été créée par un agent IA (OpenHands) pour le compte de @kadohack12.

@kadohack12 can click here to continue refining the PR

Adds CI so every push and PR is validated automatically. Previously these
repos had tests (or a build) but no pipeline running them.

Co-authored-by: openhands <openhands@all-hands.dev>
@kadohack12
kadohack12 marked this pull request as ready for review October 31, 2026 17:33
@kadohack12
kadohack12 merged commit a45aba3 into main Oct 31, 2026
3 checks passed
@kadohack12
kadohack12 deleted the ci/github-actions branch October 31, 2026 17:35
kadohack12 added a commit that referenced this pull request Nov 15, 2026
Add sliver_run_discovery(session_id, command), an adversary-emulation
helper that audits a target using a strict read-only allow-list (whoami,
id, uname -a, hostname, uptime). Anything else is refused before it
reaches the target.

- sliver.py: allow-list, session-id validation, and clear refusals
- core.py: expose sliver_run_discovery to the model via TOOL_SCHEMAS
- cli.py: classify it state-changing (deny-by-default [y/N]), support
  model-supplied tool arguments, and add a 'discover' CLI command
- tests: allow-list enforcement, argument routing, and an end-to-end
  test where 'Qui est l'utilisateur sur la session ?' runs whoami through
  the Safety Gate and returns the model's analysed answer
- docs: README (features, safety model, Role #2/#3, CLI) and
  WALKTHROUGH (optional T1082 step); test badge 102 -> 132

Co-authored-by: openhands <openhands@all-hands.dev>
kadohack12 added a commit that referenced this pull request Nov 15, 2026
Adds CI so every push and PR is validated automatically. Previously these
repos had tests (or a build) but no pipeline running them.

Co-authored-by: openhands <openhands@all-hands.dev>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants