Claude (review-harness setup) f3ee4722a8 Phase A + B (MVP) — local review harness
Implements the MVP cutline from the planning artifact:
- Phase A: skeleton + CLI dispatch + provider interface + stub model doctor
- Phase B: scanner + git probe + 12 static analyzers + reporters + pipeline
- Phase B fixtures: clean-repo, insecure-repo, degraded-repo

12 static analyzers per PROMPT.md "Suggested Static Checks For MVP":
hardcoded_paths, shell_execution, raw_sql_interpolation, broad_cors,
secret_patterns, large_files, todo_comments, missing_tests,
env_file_committed, unsafe_file_io, exposed_mutation_endpoint,
hardcoded_local_ip.

Acceptance gates passing:
- B1 (intake produces accurate counts) ✓
- B2 (insecure fixture fires ≥8 distinct check_ids — actually 11/12) ✓
- B3 (clean fixture produces 0 confirmed findings — no false positives) ✓
- B4 (scrum mode produces all 6 required markdown + JSON reports) ✓
- B5 (receipts.json marks degraded phases honestly) ✓
- F  (self-review on this repo runs without crashing) ✓ — exit 66 (degraded
  because Phase C LLM review is hardcoded skipped)

Phases C (LLM review), D (validation cross-check), E (memory + diff +
rules subcommands) deferred per the cutline. The MVP delivers the
evidence-first path; LLM is purely additive.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-30 00:56:02 -05:00

53 lines
2.0 KiB
Go

// Package llm defines the model-provider abstraction. Phase A ships
// the interface only; Phase C adds the Ollama implementation.
//
// Provider interface mirrors PROMPT.md / docs/LOCAL_MODEL_SETUP.md:
// list_models()
// complete(prompt, options)
// complete_json(prompt, schema, options)
// health_check()
//
// Phase A's stub doctor uses this only for HealthCheck — the rest
// is wired in Phase C.
package llm
import "context"
// HealthStatus is what HealthCheck returns. Stable shape so the
// model-doctor JSON schema doesn't shift between phases.
type HealthStatus struct {
ServerAvailable bool `json:"server_available"`
PrimaryModelAvailable bool `json:"primary_model_available"`
FallbackModelAvailable bool `json:"fallback_model_available"`
BasicPromptOK bool `json:"basic_prompt_ok"`
JSONModeOK bool `json:"json_mode_ok"`
Errors []string `json:"errors"`
}
// CompleteOptions tunes a non-streaming completion call.
type CompleteOptions struct {
Temperature float64
MaxTokens int
TimeoutSeconds int
}
// Provider is the abstraction every model backend implements.
// G0 ships Ollama; OpenAI-compatible local endpoints land in
// Phase F+ when the harness needs them.
type Provider interface {
// Name returns the short identifier (e.g. "ollama").
Name() string
// HealthCheck probes server + primary + fallback model availability
// + a basic prompt + a JSON-mode probe. Used by `model doctor`.
HealthCheck(ctx context.Context, primaryModel, fallbackModel string) HealthStatus
// Complete performs a non-streaming completion. Phase C wires this.
Complete(ctx context.Context, model, prompt string, opts CompleteOptions) (string, error)
// CompleteJSON requests strict JSON output. Phase C wires this.
// Implementations should set Ollama's `format: "json"` or its
// upstream-equivalent constrained-decoding flag.
CompleteJSON(ctx context.Context, model, prompt string, opts CompleteOptions) (string, error)
}