feat(providers): add gemini provider (Google Gemini) - #129
Merged
Conversation
- Added 'gemini' to _VALID_PROVIDER_MODULES in loader.py - Registered in KNOWN_PROVIDERS, PROVIDER_CATALOG, and PROVIDER_CREDENTIAL_VARS - Uses standard GOOGLE_API_KEY credential path (no custom branch needed) - Added to bundle.md provider stub list Generated with [Amplifier](https://github.com/microsoft/amplifier) Co-Authored-By: Amplifier <240397093+microsoft-amplifier@users.noreply.github.com>
- Updated README with gemini provider listing - Added CHANGELOG entry for provider addition - Updated CONFIGURATION.md with gemini config example - Updated providers-and-models.md specification with Gemini details - Updated host-config.md credential documentation - Updated bundle-and-cache.md provider reference - Updated LAYERS_AND_RELEASES.md version notes - Updated INTEGRATION.md with gemini integration info - Updated architecture.dot diagram to reflect new provider - Updated SKILL.md with gemini provider documentation Generated with [Amplifier](https://github.com/microsoft/amplifier) Co-Authored-By: Amplifier <240397093+microsoft-amplifier@users.noreply.github.com>
Covers the catalog registration that made gemini the eighth supported provider: `providers list` reports it against the provider-gemini module and GOOGLE_API_KEY, `models list --provider gemini` returns a live listing, a host config naming it passes validation, and a session completes against it. Forward GOOGLE_API_KEY through the e2e DTU profile so the suite can reach Google. GOOGLE_API_KEY is the canonical variable: the Google GenAI SDK also accepts GEMINI_API_KEY, but GOOGLE_API_KEY takes precedence and is the sole entry in PROVIDER_CREDENTIAL_VARS, so it is what `providers list` and `models list` actually consult. The suite skips rather than fails when the credential is absent from the container, unlike github_copilot which fails loud. Its cases guard the shared provider catalog, so it runs on every full `cli.py run`; failing the default run for a contributor with no Google key would be red for a reason unrelated to their change. Note: with a Google key present, `serve` auto-enables gemini from resolvable credentials, so the shared in-DTU HTTP server lists gemini models alongside anthropic. Existing HTTP suites resolve a model id at runtime and are unaffected. 🤖 Generated with [Amplifier](https://github.com/microsoft/amplifier) Co-Authored-By: Amplifier <240397093+microsoft-amplifier@users.noreply.github.com>
# Conflicts: # CHANGELOG.md
David Koleczek (DavidKoleczek)
added a commit
that referenced
this pull request
Aug 19, 2026
main added gemini as the eighth provider (#129) while this branch added vllm as the eighth. Both are now registered, making nine, and every "eight providers" reference has been updated to match. Conflicts resolved by keeping both providers in each list: - provider_sources.py: KNOWN_PROVIDERS and PROVIDER_CATALOG (verified in sync) - config/loader.py: _VALID_PROVIDER_MODULES, reflowed to one name per line - bundle/bundle.md: both install-only provider stubs - CHANGELOG.md: the vllm entry moved from [0.13.0] to [Unreleased], since 0.13.0 was cut on main before this branch's work landed - README.md, LAYERS_AND_RELEASES.md, architecture.dot, SKILL.md, INTEGRATION.md, spec/host-config.md, spec/bundle-and-cache.md, spec/providers-and-models.md: provider counts and lists 🤖 Generated with [Amplifier](https://github.com/microsoft/amplifier) Co-Authored-By: Amplifier <240397093+microsoft-amplifier@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds
geminias a first-class provider in amplifier-agent's provider catalog, wiring up the existing official moduleamplifier-module-provider-gemini— Google Gemini via the Google AI API. It is a normal API-key provider (likeanthropic/openai). Listed in the ecosystemMODULES.md; this completes another gap in amplifier-agent's hardcoded provider set (sibling to #127 / #128).What changed
Code
config/loader.py: addgeminito_VALID_PROVIDER_MODULES.provider_sources.py: add toKNOWN_PROVIDERS,PROVIDER_CATALOG, andPROVIDER_CREDENTIAL_VARS(registered asGOOGLE_API_KEY). The module also acceptsGEMINI_API_KEY; only the primary is listed to avoid a spurious deprecation notice (same pattern asgithub-copilot). Uses the existing generic credential path — no dedicated branch, andauth set geminiworks normally.bundle.md: declareprovider-geminiin the install-only stub list.Docs: README,
docs/spec/*,docs/CONFIGURATION.md,docs/LAYERS_AND_RELEASES.md,docs/INTEGRATION.md,docs/architecture/architecture.dot,skills/amplifier-agent/SKILL.md,CHANGELOG.md.Also incidentally corrects two pre-existing stale provider counts in
docs/spec/host-config.md("six valid module names", "outside the 5 valid names") that had fallen behind earlier catalog merges — both now reflect the true current count.Verification (Digital Twin, real Google Gemini — no mocks)
Verified end-to-end in an isolated container running the patched build (
amplifier-agent 0.12.0) against the real Google AI API (gemini-2.5-flash). The API key was handled by reference only and never left the container.providers list:geminiappears;resolvable=false/source=nonewithGOOGLE_API_KEYunset, flips toresolvable=true/source=envwhen set.provider.moduleis rejected with a valid set that now includesgemini.amplifier-agent runreturned a correctgemini-2.5-flashcompletion ("gemini live") with a clean terminated envelope (provider=gemini,model=gemini-2.5-flash).generativelanguage.googleapis.commade the same turn fail with a connect error whose traceback runs throughprovider-gemini; restoring it made it succeed. No fallback was possible (onlygeminiresolvable).bundle.md; the prepared bundle declaresprovider-gemini.