gencersarp/Agent Sandbox

Run LLM agents safely inside CI with strict capability manifests, resource limits, and human approval.

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Activelast commit May 21, 2026
License
None

Pinned Snippet

workflow.ymlSHA-pinned
uses: gencersarp/agent-sandbox@489f025ffe8401271599f2fdee9f84d76233d9af # no releases — HEAD as of 2026-07-11

tags can be moved; commit SHAs can't. why a SHA?

namedescriptionrequireddefault
manifest_pathPath to the .agent-sandbox.yml manifest file relative to the repo root.no.agent-sandbox.yml
llm_api_keyAPI key for the LLM endpoint (pass via secrets).yes
llm_api_urlOverride the LLM API URL (default: OpenAI chat completions).nohttps://api.openai.com/v1/chat/completions
llm_modelLLM model to use.nogpt-4o
create_prWhether to create a draft PR from the agent's changes (true/false).nofalse
base_branchBase branch for the PR and diff context.nomain
dry_runWhether to run in dry run mode (true/false). Shows changes but does not generate patches/PRs.nofalse
namedescription
report_pathPath to the JSON report file.
patch_pathPath to the unified diff patch file.
pr_urlURL of the created draft PR (empty if create_pr is false or no changes).