hydralisted

<!-- v2.0: per ADR docs/adr/0001-execution-substrate.md (Option C, accepted 2026-05-23), interactive /hydra runs in THIS harness (Agent tool + codex-companion), so SKILL.md is the living product surface. Prompt-level v2.0 wins land here directly (Echo advisor; chairman grounding + suspicious-verdict gate). The Python core in hydra/ is scoped to the bench (deterministic citation-grounding + reproducible scoring) -- it does NOT replace this runtime. Design context: docs/specs/2026-04-17-hydra-2.0-core-design-v2.md. --> # Hydra ## Example invocation ```text Run Hydra for a deep security and architecture review of this branch. ``` Four advisors analyze your code from different angles by default (standard mode) -- including Echo, which reviews AI-assisted-development failure modes. Escalate to deep mode for the full council: six advisors, three cross-examining reviewers, and a chairman synthesizing the final verdict. Standard mode runs 4 advisors + chairman on Opus (~$0.35-0.65). Deep mode adds 2 more advisors (including 2 Codex), 3 reviewers, and cross-model diversity (~$1.50-2.50). Reference files in `references/` define all prompts and protocols -- read them at the relevant step. --- ## Modes | Mode | CLI | Advisors | Reviewers | Chairman | Total | Est. Cost | |------|-----|----------|-----------|----------|-------|-----------| | **standard** | *(default)* | 4 (Cassandra + Mies+ + Sentinel + Echo) | 0 | 1 Opus | 5 | ~$0.35-0.65 | | **deep** | `