v0.2.0

[0.2.0] — 2026-07-13

Changed

• Hawk/Eyrie production boundary completed: Hawk owns the product face,
  sessions, tools, permissions, and public schemas while Eyrie v0.2.1 owns
  credentials, catalog resolution, provider transport, resilience, and usage
  telemetry behind the stable eyrie/engine facade.
• Provider routing and usage attribution hardened: resolved route changes,
  continuation segments, and terminal usage are propagated without duplicate
  accounting, and production Eyrie calls use exactly one resilience layer.
• Daemon conversations are durable: JSON and SSE chat requests create or
  resume persisted sessions, expose stable session IDs, preserve metadata, and
  distinguish invalid, missing, and corrupt state.
• Release and supply-chain gates strengthened: exact ecosystem Gitlinks,
  module/tag parity, Trivy enforcement, public-module builds, SBOM generation,
  and cross-platform artifacts are part of the release path.
• Fixed /mode auto misclassifying plain English as shell commands: shellmode.ClassifyInput trusted exec.LookPath(firstWord) alone, so any sentence starting with a word that's also a real Unix binary (make sure this works, find the bug in this file, kill the old branch, sort out the imports...) was silently executed as a shell command instead of being sent to the model — confirmed 18 of 20 sampled sentences misclassified before the fix. Now a curated allowlist of unambiguous dev-tool names (git, npm, docker, ls, cat, ...) is trusted immediately, while every other PATH match requires real shell-syntax evidence (a flag, a path, a file extension, or an operator like |/>/&&) before being trusted as a shell command — matching the same ambiguity Warp's own terminal autodetect documents and resolves with a user-configurable denylist.
• Permission system unified into two independent axes: the old PermissionMode (default/acceptEdits/bypassPermissions/dontAsk/plan) is removed. /autonomy now controls the 5-tier trust ladder (Always Ask/Scout/Builder/Operator/Autonomous, bare /autonomy opens a picker), and /spec controls an independent, orthogonal spec-driven workflow gate (Specify → Plan → Tasks → ApproveImplementation, bare /spec opens a picker) that blocks Write/Edit/Bash regardless of trust tier — including at Autonomous. Fixes a real bug where the old Plan Mode's write-block could be silently bypassed at high autonomy tiers, since tier and mode were checked independently with no ordering guarantee.
• Fixed PermissionService.SetAutonomy/Autonomy(): previously wrote to/read from a shadow field the permission engine's CheckTool never consulted, meaning autonomy tier changes may not have reliably taken effect. Now both read/write the same PermissionEngine.Autonomy field the check logic uses.
• --permission-mode CLI flag removed; --dangerously-skip-permissions unchanged (now maps to the Autonomous tier). New --dry-run flag added as an unconditional kill switch (deny every tool call, regardless of tier or spec stage) — replaces dontAsk's hard-lockout role.
• Version re-baselined to 0.1.0 across cmd/hawk/main.go, cmd/daemon.go,
  flake.nix, .github/workflows/release.yml, and the update/daemon test suites, aligning hawk
  with the rest of the GrayCodeAI ecosystem (eyrie, tok, yaad, sight, inspect).
• Architecture boundary hardening: Hawk now owns runtime request/response DTOs, transport config/provider seams, and review/verification product-boundary contracts, with eyrie/client usage restricted to internal adapters and guarded in CI.
• shared/types removed: Hawk no longer ships the old shared type path, and local boundary checks now block any attempt to reintroduce it.

Added

• Spec-driven workflow (/spec): independent, orthogonal permission gate that walks the model through Specify → Plan → Tasks, writing real spec.md/plan.md/tasks.md files to .hawk/specs/<slug>/, and requires explicit ApproveImplementation approval (always prompts, at any trust tier) before Write/Edit/Bash unlock. The approval prompt shows the actual written content, not a blind yes/no.
• /autonomy and /spec picker overlays: bare /autonomy or /spec opens an arrow-key-navigable, filterable picker (Esc/Enter) instead of requiring subcommand syntax; typed subcommands (/autonomy tier scout, /spec status, etc.) still work.
• Watch mode (--watch): file-watcher loop that acts on AI! (do-now) and AI? (answer) code comments. Off by default.
• GitHub Action (.github/actions/hawk): interactive mode on @hawk mentions, automation mode on labeled issues/PRs, and skill dispatch for /-prefixed prompts.
• Messaging gateways: opt-in Telegram, Discord, and Slack gateways on the daemon for chatting with hawk from messaging apps.
• AST repo-map (internal/context/repomap): structural repository map for richer model context.
• Auto codebase analysis on first run (internal/autoinit): opt-in seeding of project context.
• Auto-lint / auto-fix cycle: runs the matching linter after edits and iterates on fixes with bounded retries (opt-in).
• Image / multimodal context (internal/engine/vision.go): feed screenshots/images to vision-capable models.
• Plan & Explore sub-agent modes: read-only plan (task decomposition) and explore (codebase investigation) modes; explore supports quick/medium/very-thorough thoroughness budgets.
• Persona color and hooks: per-agent display color and lifecycle hook (pre_run/post_run) fields.
• YAML agents & eval tasks: define personas and eval tasks in YAML alongside markdown personas.
• IT-managed policy tier (internal/context/rules.go): non-excludable org-policy rule tier with highest precedence; strips HTML comments from rule files.
• SQL exploration tool (internal/tool/sql.go): read-only database exploration tool.
• Conventional-commit generation: SmartCommit and the diff summarizer produce Conventional Commit messages from staged changes.
• Durable workflows + named checkpoints (internal/multiagent): LangGraph-style resumable workflows with named, persisted step boundaries.
• Human-in-the-loop approval gates (internal/engine/approval_gate.go): durable approve/reject gates within workflows.
• Structured output (internal/engine/structured_output.go): JSON-Schema-constrained responses, validated and retried once on mismatch.
• MCP WebSocket transport (internal/mcp/ws.go): opt-in WebSocket transport in addition to stdio/HTTP.
• GET /v1/ready: dependency-aware readiness endpoint on the daemon.
• REPL magic commands (%reset, %undo, %tokens, %history, %copy, %save, %compact, %model, %clear)
• Prompt cache keep-alive pings
• Unified finding/severity contracts now live in hawk-core-contracts/types

Added — Round 2 ecosystem improvements (2026-06-01)

• Cavecrew personas (internal/multiagent/agents): three new
  built-in personas built into GrayCode Hawk
  (cavecrew-investigator, cavecrew-builder, cavecrew-reviewer).
  Each enforces a strict output format so downstream agents can parse
  outputs mechanically:
  • cavecrew-investigator: path:line — symbol — note, max 6 words per note
  • cavecrew-builder: hard-refuses tasks touching 3+ files
  • cavecrew-reviewer: severity emoji (🔴/🟡/🔵/❓) at the start of every line
    Exposed via CavecrewPersonas() helper and EnsureCavecrew()
    registry method; BuiltinPersonas() returns 21 (was 18).
• internal/safewrite package: hardened atomic file-write utility
  native GrayCode hardened atomic writes. Refuses symlinks at destination
  and parent, refuses paths that escape via .., opens with
  O_NOFOLLOW via golang.org/x/sys/unix, writes to a temp file
  with mode 0600, syncs to disk, then atomically renames. ErrSymlinkTarget
  and ErrPathEscape sentinel errors.
• internal/jsonc package: JSON-with-Comments parser and
  ValidateClaudeSettings validator, native GrayCode settings
  parser and validateHookFields. Accepts // and /* */ comments
  plus trailing commas in objects and arrays. Validates Claude Code
  settings.json fields (model, permissions, hooks, mcpServers,
  env) with type checks and value validation.
• internal/permissions/verdict.go: unified PermissionVerdict
  type with Risk levels (RiskLow, RiskMedium, RiskHigh,
  RiskBlocked). Helpers: Allow, Deny, RequireApproval. Additive
  change — existing GuardianDecision is unchanged.
• internal/providers package: PROVIDERS matrix (34 entries)
  native GrayCode PROVIDERS matrix. Each
  entry describes an AI coding agent (Claude Code, Cursor, Codex,
  Aider, etc.) with install mechanism and detection probes. Probe
  kinds: command (PATH), dir (filesystem), vscode-ext,
  cursor-ext, macapp, jetbrains-plugin. API: Get(id),
  All(), Hard(), Detect(). Soft: true means detection is
  best-effort; soft providers are excluded from auto-detect.
• internal/session.GainTracker: per-session gain event recording
  in a new gains SQLite table inside the session store. Each event
  captures original/compressed byte + token counts, mode/tier/model,
  and a command label. API: Record, AggregateForSession,
  ListForSession, PruneForSession. Scoped to a single session so
  callers can selectively compact their own history. Companion to
  tok's internal/tracking.Tracker (tok tracks globally, hawk tracks
  per-session).

Added — Production Hardening (top-50 OSS parity)

• Stricter linting: .golangci.yml v2 config enabling errcheck, staticcheck,
  gocritic (diagnostic + performance), unused, ineffassign, misspell, noctx,
  bodyclose, unconvert, whitespace, with govet enable-all (minus fieldalignment).
• CI parity: race-detector tests with coverage upload, golangci-lint v2 action,
  `g...

v0.1.0

Changelog

• 9ed5605 Add cmdhistory: structured command history with FTS5 search
• 030b315 Add convodag: conversation-as-DAG with SQLite persistence
• 0b48ecd Add cost optimization: wasted spend detection + model downgrade suggestions
• eeb7648 Add diffsandbox: virtual file overlay for proposed edits
• 420dbd4 Add fingerprint: fast repo summary for LLM context injection
• f4f224c Add incremental repomap + structured planner
• 806073c Add localize: hierarchical fault localization (file → symbol → edit)
• 25a408c Add parallel: git worktree-based parallel task execution
• 23803cb Add rules: import/export between AI tool formats
• 9fcbe40 Add session trace trees: parent-child span hierarchy with tree rendering
• 95af523 Add trace to ecosystem references
• 75da55a Add world-class solo developer ecosystem features
• e73d2fe Fix 3 lint issues: ineffassign + staticcheck
• 02ce156 Fix CI: add workspace + dep checkout to all jobs (test, lint, build)
• 23623f5 Fix CI: goinstall mode for golangci-lint, fix plan.go vet warning
• 8af3d0e Fix CI: golangci-lint v2 config + git branch test fix
• f30c7e1 Fix CI: update dependency versions, fix golangci-lint compatibility
• ea05e26 Fix CI: use workspace with cloned deps, upgrade to golangci-lint-action v7
• 7545b0a Fix core gaps: per-tool timeout, permission/session persistence, usage streaming
• aaf6433 Fix golangci-lint v2 config format
• 38853ad Fix lint: add default:none to disable errcheck
• 344d615 Fix lint: disable errcheck (too noisy for existing codebase)
• b4b02a3 Fix lint: keep only govet+ineffassign (unused has too many pre-existing hits)
• 2114c02 Fix lint: keep only govet+ineffassign+unused (no pre-existing noise)
• 71384b6 Fix lint: remove gosimple (merged into staticcheck in v2)
• 9d6e6bf Fix module path: use github.com/GrayCodeAI/eyrie everywhere (matches repo URL)
• 8ef32b0 Integrate trace: add hawk capture command for Git-native session recording
• 924464b Merge branch 'dev'
• e1f4002 Merge pull request #1 from GrayCodeAI/dev
• 17a35e9 Use hawk trace as primary command name
• 3435027 ci: add GitHub Actions workflows for test, build, lint, and release
• ddd7e4e ci: fix GOWORK path for GoReleaser
• 825366e ci: fix release workflow to clone sister repos for go.work
• d41b0b8 ci: remove go mod tidy from goreleaser hooks (uses go.work)
• 11bbf56 ci: trigger on dev branch
• 7f410d6 ci: trigger on dev branch
• 5573e52 ci: use go build instead of test in goreleaser hook (tests run separately)
• 14cb8fa feat(tui): improve spinner, error handling, and UI polish
• ea2795f feat: 100% parity — config, plan mode, smart compaction, expanded prompt, 17 tools
• 532b273 feat: BOM decoding, include/exclude patterns, nested gitignore, better MCP descriptions
• eb63680 feat: Phase 1 — core agent loop with eyrie streaming
• 1e36e25 feat: Phase 2 — tool system + agentic loop
• 358be64 feat: Phase 3 — REPL polish, sessions, slash commands, context
• 9fae8a1 feat: Phase 4-6 — resume, cancel, slash commands, README
• b4ce723 feat: UX improvements inspired by Droid/Kiro CLI
• 010a65c feat: activate convodag for conversation branching
• feb5a44 feat: add /branches command and register in slash command list
• 2d9814a feat: add /metrics slash command to TUI
• d387abd feat: add 12 frontier features from top OSS coding agents
• c677643 feat: add 4 competitive OSS features — universal provider, sandbox, repo map, git auto-commit
• 31cf908 feat: add 8 Plandex-inspired workflow features
• 0ce1f76 feat: add 9 Karpathy-inspired features — vibe mode, research loop, power dial
• bf33cc5 feat: add Docker container mode (herm-style hermetic execution)
• a8d107e feat: add Docker support with multi-stage build
• 66a2a14 feat: add Nix flake for reproducible builds
• 30928a3 feat: add cascade routing and full ecosystem integration tests
• 957642f feat: add circuit breaker package for API resilience
• 13a7780 feat: add config validation, benchmarks, and fuzz tests
• d55a3b6 feat: add graceful shutdown package
• f354f7a feat: add herm-inspired modular prompts, workspace context, sub-agent budgets
• 29b79eb feat: add logger, health, and metrics packages
• 2f719f5 feat: add profiling and tracing packages
• 3ac813e feat: add rate limiter with token bucket algorithm
• bd7e1ae feat: add retry package with exponential backoff for API resilience
• f3f8e75 feat: add shell completion and golangci-lint config
• aa1f427 feat: add viewport scrolling, alt screen, and proper layout
• aa79089 feat: adopt cocoindex approach — incremental indexing via tok chunking + yaad storage
• f918ddd feat: auto-build hawk image on first run (no fallback to bare ubuntu)
• 6582b27 feat: block input completely when Docker not running (like herm)
• e5a09e0 feat: complete all 225 TODO items — full solo-dev feature set
• 960226e feat: complete archive parity - hooks, plugins, permissions, models, memory, analytics, auth, update, lsp, voice, magicdocs, sandbox, remote, ide
• 17c3d40 feat: complete archive parity migration — advanced compaction, OAuth, OTel, vim, routing, bridge, and more
• 2a672dd feat: container-only mode with Go pre-installed (no fallback)
• b1332e5 feat: enforce hawk identity across all LLM providers
• 4299d53 feat: expand repomap to 20 languages, extract PermissionEngine
• f1da4bd feat: full herm-style container mode with TUI status bar
• 55ab3d6 feat: full parity — 12 tools, permission memory, agent spawning, slash commands
• 4c8e226 feat: harden session safety — atomic writes, WAL crash recovery, corruption tolerance
• 67c1ed9 feat: hawk Go CLI scaffold — cobra + bubbletea + eyrie
• b589c7a feat: implement 10 research-grade innovations from 2025-2026 papers
• 217cc65 feat: implement 100+ TODO items — error handling, markdown, safety, config, sessions
• bfb33b0 feat: implement 14 features from herm/langdag comparison
• 9aee44b feat: implement Droid-parity features — Ctrl+N/L shortcuts, /new, /btw, /retry, /add, /run, /test, /lint
• b71e01e feat: implement remaining TODO items — performance, DX, TUI, config, testing
• 563e857 feat: integrate health checks into doctor command
• e16427b feat: integrate logger and metrics into engine
• c665b0e feat: integrate tok (token counting) and yaad (memory graph) into hawk
• 65b4cae feat: major parity push — add 15 new tools, swarm coordination, 30+ slash commands, remote bridge
• deb36c1 feat: make container mode default (auto-detect Docker like herm)
• 380f563 feat: match herm container setup (version-tagged image, same-path mount, attachments/cache)
• 818b605 feat: parallel indexing, BM25 re-ranking, CodeSearch tool
• c4ad670 feat: port final 13 solo-dev features from hawk-archive
• acf5696 feat: production-grade — MCP, goreleaser, ask_user, tests, README
• 363e0a7 feat: render markdown bold as ANSI bold, fix identity regex
• f654757 feat: v0.4.0 — exec, daemon, mission, agents, snapshots, Waza workfl...