Answer: Contributors should follow a seven-step workflow: identify the target behavior, author the eval file under `evals/.eval.ts`, configure workspace files, assert behavior not prose, run locally with `RUN_EVALS=true npx vitest run evals/my-test.eval.ts`, deflake by running at least 3 times, and run `npm run eval:validate`. Acceptance criteria require the file to end with `.eval.ts` or `.eval.tsx`, new evals to start as `USUALLY_PASSES`, static `suiteName` and `suiteType` specified, assertions using `rig.waitForToolCall` or explicit tool arguments, and no writes outside `rig.testDir`. Status: Confirmed Entry: tool/gemini-cli/agents-and-automation/contributor-workflow-acceptance Revision: 1 Digest: sha256:541d5a1883a8f729 (changes whenever the content changes) Timeline: Created: 2026-11-13T03:17:15Z Last changed: 2026-11-13T03:17:15Z Last verified: 2026-11-20T12:50:30Z Next check: 2026-11-21T12:50:30Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Source | As of eval file naming | evals/.eval.ts | [1] | 2026-11-20 local run command | RUN_EVALS=true npx vitest run evals/my-test.eval.ts | [1] | 2026-11-20 deflake runs | at least 3 times | [1] | 2026-11-20 validation command | npm run eval:validate | [1] | 2026-11-20 file extension | .eval.ts` or `.eval.tsx | [1] | 2026-11-20 initial policy | USUALLY_PASSES | [1] | 2026-11-20 suite type | 'behavioral' | [1] | 2026-11-20 workspace restriction | rig.testDir | [1] | 2026-11-20 Sources: [1] Google. "Behavioral Evaluations & EDK Guide". https://geminicli.com/docs/behavioral-evals.md Role: primary. Accessed: 2026-11-20T12:50:30Z Quote: "Create your file under `evals/.eval.ts`" (Supports: eval file naming) Quote: "RUN_EVALS=true npx vitest run evals/my-test.eval.ts" (Supports: local run command) Quote: "Run your eval at least 3 times locally to verify it does not" (Supports: deflake runs) Quote: "Run `npm run eval:validate` to ensure no linting errors" (Supports: validation command) Quote: "File ends with `.eval.ts` or `.eval.tsx`." (Supports: file extension) Quote: "New evals start as `USUALLY_PASSES`." (Supports: initial policy) Quote: "Static `suiteName` and `suiteType` (e.g. `'behavioral'`) are" (Supports: suite type) Quote: "Does not write to files outside `rig.testDir`." (Supports: workspace restriction) Published: 2026-10-16T02:36:07Z Snapshot SHA-256: 2124cd48c1c85f7a5401a6ba1094ccd5741dc6eedcdadc069b6e7c3a39a0e49e (captured 2026-11-20T12:50:27Z) Verification: Method: Automated extraction from [1] with deepseek-flash (JSON output, thinking disabled). Every value was checked by code to appear verbatim in a quoted passage of [1]. On this check the text of [1] was unchanged since 2026-11-13T03:17:15Z, so the values verified then were kept. Cross-check: not-available at 2026-11-20T12:50:30Z. No independent second source is configured for this entry Last successful check: 2026-11-20T12:50:30Z Last attempt: 2026-11-20T12:50:30Z Notes and limitations: Extracted automatically from a single official source (section "Contributor Workflow"). Related: https://latestn.com/tool/gemini-cli/agents-and-automation/create-custom-agent-skill How do I create a custom Agent Skill in Gemini CLI? https://latestn.com/tool/gemini-cli/agents-and-automation/create-custom-subagent How do I create a custom subagent in Gemini CLI? https://latestn.com/tool/gemini-cli/agents-and-automation/create-hook-script How do I create a hook script for Gemini CLI? https://latestn.com/tool/gemini-cli/agents-and-automation/activate-skill-purpose What does the activate_skill tool do in Gemini CLI? https://latestn.com/tool/gemini-cli/agents-and-automation/ask-for-task-plan How do I ask Gemini CLI to create a task plan for a complex job? https://latestn.com/tool/gemini-cli/agents-and-automation/automated-issue-triage What automated triage happens when I open an issue in Gemini CLI? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-13T03:17:15Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/gemini-cli/agents-and-automation/contributor-workflow-acceptance/latest Challenge this entry: Open challenges: None How: If a value here is wrong or out of date, send what is wrong and a link to evidence. Challenges are queued and checked against the evidence; they do not change the entry by themselves. If the entry changes as a result, the change log says so. Form: https://latestn.com/dispute/tool/gemini-cli/agents-and-automation/contributor-workflow-acceptance POST: https://latestn.com/dispute/tool/gemini-cli/agents-and-automation/contributor-workflow-acceptance with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/gemini-cli/agents-and-automation/contributor-workflow-acceptance?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). What is the contributor workflow and acceptance criteria for adding behavioral evaluations to Gemini CLI? Revision 1, last verified 2026-11-20T12:50:30Z. https://latestn.com/tool/gemini-cli/agents-and-automation/contributor-workflow-acceptance (accessed 2026-11-21T10:01:47Z). @misc{LatestN_tool_gemini_cli_agents_and_automation_contributor_workflow_acceptance_r1, title = {What is the contributor workflow and acceptance criteria for adding behavioral evaluations to Gemini CLI?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/gemini-cli/agents-and-automation/contributor-workflow-acceptance}}, note = {Revision 1; digest sha256:541d5a1883a8f729; last verified 2026-11-20T12:50:30Z; accessed 2026-11-21T10:01:47Z} } Reuse: License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/) Attribution: Copying and adapting are allowed, including commercially, if you credit LatestN and link to the entry you used. Each entry ends with a ready-made citation. URL: https://latestn.com/tool/gemini-cli/agents-and-automation/contributor-workflow-acceptance