fix: correct statusline context calculation and track hooks in manifest (#769)

CHANGELOG.md

@@ -6,6 +6,41 @@ Format follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 
 ## [Unreleased]
 
+### Fixed
+- **Statusline context bar** — corrected calculation to use Claude Code's actual 16.5% autocompact buffer instead of a hardcoded 80% ceiling, so the displayed percentage now reflects when compaction is truly imminent (#769)
+  - Color thresholds updated to intuitive levels: green < 50%, yellow < 65%, orange < 80%, red ≥ 80%
+  - Display now shows two decimal places (e.g. `42.37%`) so small context changes are visible
+- **Hooks not tracked in file manifest** — `gsd-statusline.js`, `gsd-check-update.js`, and `gsd-context-monitor.js` are now included in `gsd-file-manifest.json` so local modifications are detected and backed up before `/gsd:update` (#769)
+- **Quality/balanced profiles now deliver Opus subagents** — `resolveModelInternal` previously
+  converted `opus` to `inherit`, causing agents to silently run on Sonnet when the parent
+  session used the default Sonnet 4.6 model. Opus is now passed directly to Task calls (#695)
+
+### Added
+- **Kimi CLI support** — install GSD skills and agents to [Kimi](https://github.com/MoonshotAI/kimi-cli) via `--kimi --global`
+  - Skills install to XDG path `~/.config/agents/skills/gsd-<name>/SKILL.md` and are invoked with `/skill:gsd-<name>`
+  - Agents install as dual-file format: `~/.kimi/agents/gsd-<name>.yaml` + `gsd-<name>.md` system prompt
+  - Tool names translated to full module paths (`kimi_cli.tools.file:ReadFile`, `kimi_cli.tools.shell:Shell`, etc.)
+  - MCP tools excluded from agent tool lists (configured separately via Kimi's config)
+  - `${VAR}` patterns in agent bodies escaped for Kimi CLI compatibility
+  - Local install guard — Kimi's XDG skills path is always global; `--kimi --local` exits with a clear error
+  - `KIMI_CONFIG_DIR` and `KIMI_SKILLS_DIR` env overrides for custom paths
+- **Adaptive model profile** — fourth model profile (`adaptive`) that auto-selects models per-plan based on complexity evaluation (#210)
+  - `evaluateComplexity()` scores plan metadata (files modified, task count, objective keywords, plan type, dependencies) on 0-10+ scale
+  - Three tiers: Simple (haiku/sonnet), Medium (sonnet/opus), Complex (opus/sonnet)
+  - `adaptive_settings` config: `min_model`/`max_model` clamping, `log_selections` usage logging
+  - `resolve-adaptive-model` CLI command with `--context` for per-plan resolution
+  - `init plan-phase` and `init quick` now return `model_profile` and `adaptive_settings` for workflow use
+  - Plan index includes `type` and `depends_on` for enriched complexity evaluation
+  - Verifier uses per-plan adaptive resolution when in adaptive mode
+  - Full backward compatibility — non-adaptive profiles unaffected
+- `/gsd:report-bug` command for structured bug reporting with severity tracking, diagnostic log capture, and GitHub issue creation
+  - `gsd-tools bug list/update/resolve` CLI commands for bug management
+  - `gsd-tools init bugs` and `scaffold bugs` for workflow bootstrapping
+  - Severity inference from keywords (critical/high/medium/low)
+  - Automatic diagnostic capture (git state, log files, error output)
+  - Optional GitHub issue creation via `gh` CLI
+  - Bug lifecycle: reported → investigating → fixing → resolved
+
 ### Fixed
 - Scope phase counting in `buildStateFrontmatter` and `cmdPhaseComplete` to current milestone — multi-milestone projects no longer report inflated total/completed phases
 - Use ROADMAP phase count for `total_phases` when phases lack directories — prevents premature milestone completion detection
@@ -25,6 +60,14 @@ Format follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 - Analysis paralysis guard in agents to prevent over-deliberation during planning
 - Exhaustive cross-check and task-level TDD patterns in agent workflows
 - Code-aware discuss phase with codebase scouting — `/gsd:discuss-phase` now analyzes relevant source files before asking questions
+- Concurrent milestone execution: work on multiple milestones in parallel with isolated state (#291)
+  - Milestone-scoped directories under `.planning/milestones/<name>/`
+  - `ACTIVE_MILESTONE` pointer file for switching context
+  - `/gsd:switch-milestone` command with in-progress work warnings
+  - `--milestone` CLI flag for explicit milestone targeting
+  - Statusline shows active milestone in multi-milestone mode
+  - All 28 workflow files updated for milestone-aware paths
+  - Zero behavioral change for single-milestone projects (legacy mode)
 
 ### Fixed
 - Update checker clears both cache paths to prevent stale version notifications

README.md

@@ -2,7 +2,7 @@
 
 # GET SHIT DONE
 
-**A light-weight and powerful meta-prompting, context engineering and spec-driven development system for Claude Code, OpenCode, Gemini CLI, and Codex.**
+**A light-weight and powerful meta-prompting, context engineering and spec-driven development system for Claude Code, OpenCode, Gemini CLI, Codex, and Kimi.**
 
 **Solves context rot — the quality degradation that happens as Claude fills its context window.**
 
@@ -80,16 +80,18 @@ npx get-shit-done-cc@latest
 ```
 
 The installer prompts you to choose:
-1. **Runtime** — Claude Code, OpenCode, Gemini, Codex, or all
+1. **Runtime** — Claude Code, OpenCode, Gemini, Codex, Kimi, or all
 2. **Location** — Global (all projects) or local (current project only)
 
 Verify with:
 - Claude Code / Gemini: `/gsd:help`
 - OpenCode: `/gsd-help`
 - Codex: `$gsd-help`
+- Kimi: `/skill:gsd-help`
 
 > [!NOTE]
-> Codex installation uses skills (`skills/gsd-*/SKILL.md`) rather than custom prompts.
+> Codex and Kimi installations use skills (`skills/gsd-*/SKILL.md`) rather than custom prompts.
+> Kimi installs globally to `~/.config/agents/skills/` and does not support local install.
 
 ### Staying Updated
 
@@ -117,12 +119,15 @@ npx get-shit-done-cc --gemini --global   # Install to ~/.gemini/
 npx get-shit-done-cc --codex --global    # Install to ~/.codex/
 npx get-shit-done-cc --codex --local     # Install to ./.codex/
 
+# Kimi (skills-first, XDG global only)
+npx get-shit-done-cc --kimi --global     # Install to ~/.config/agents/skills/
+
 # All runtimes
 npx get-shit-done-cc --all --global      # Install to all directories
 ```
 
 Use `--global` (`-g`) or `--local` (`-l`) to skip the location prompt.
-Use `--claude`, `--opencode`, `--gemini`, `--codex`, or `--all` to skip the runtime prompt.
+Use `--claude`, `--opencode`, `--gemini`, `--codex`, `--kimi`, or `--all` to skip the runtime prompt.
 
 </details>
 
@@ -474,6 +479,7 @@ You're never locked in. The system adapts.
 | `/gsd:audit-milestone` | Verify milestone achieved its definition of done |
 | `/gsd:complete-milestone` | Archive milestone, tag release |
 | `/gsd:new-milestone [name]` | Start next version: questions → research → requirements → roadmap |
+| `/gsd:switch-milestone <name>` | Switch active milestone for concurrent work |
 
 ### Navigation
 
@@ -512,15 +518,31 @@ You're never locked in. The system adapts.
 | Command | What it does |
 |---------|--------------|
 | `/gsd:settings` | Configure model profile and workflow agents |
-| `/gsd:set-profile <profile>` | Switch model profile (quality/balanced/budget) |
+| `/gsd:set-profile <profile>` | Switch model profile (quality/balanced/budget/adaptive) |
+| `/gsd:report-bug [desc]` | Report bug with severity tracking and GitHub issue creation |
 | `/gsd:add-todo [desc]` | Capture idea for later |
 | `/gsd:check-todos` | List pending todos |
 | `/gsd:debug [desc]` | Systematic debugging with persistent state |
+| `/gsd:add-tests <N> [instructions]` | Generate unit and E2E tests for completed phase |
 | `/gsd:quick [--full]` | Execute ad-hoc task with GSD guarantees (`--full` adds plan-checking and verification) |
 | `/gsd:health [--repair]` | Validate `.planning/` directory integrity, auto-repair with `--repair` |
 
 <sup>¹ Contributed by reddit user OracleGreyBeard</sup>
 
+### Concurrent Milestones
+
+Work on multiple milestones simultaneously — e.g., v2.0 features + v1.5.1 hotfix:
+
+```
+/gsd:new-milestone "v1.5.1 Hotfix"    # Creates milestone-scoped directory
+/gsd:switch-milestone v2.0-features    # Switch back to feature work
+/gsd:progress                          # See status of active milestone
+```
+
+Each milestone gets isolated state: `STATE.md`, `ROADMAP.md`, `REQUIREMENTS.md`, `phases/` — all scoped under `.planning/milestones/<name>/`. Switch freely without losing progress.
+
+When no second milestone exists, everything stays in `.planning/` as usual (zero behavioral change).
+
 ---
 
 ## Configuration
@@ -629,9 +651,10 @@ This prevents Claude from reading these files entirely, regardless of what comma
 - Restart your runtime to reload commands/skills
 - Verify files exist in `~/.claude/commands/gsd/` (global) or `./.claude/commands/gsd/` (local)
 - For Codex, verify skills exist in `~/.codex/skills/gsd-*/SKILL.md` (global) or `./.codex/skills/gsd-*/SKILL.md` (local)
+- For Kimi, verify skills exist in `~/.config/agents/skills/gsd-*/SKILL.md`
 
 **Commands not working as expected?**
-- Run `/gsd:help` to verify installation
+- Run `/gsd:help` (or `/skill:gsd-help` for Kimi) to verify installation
 - Re-run `npx get-shit-done-cc` to reinstall
 
 **Updating to the latest version?**
@@ -656,6 +679,7 @@ To remove GSD completely:
 npx get-shit-done-cc --claude --global --uninstall
 npx get-shit-done-cc --opencode --global --uninstall
 npx get-shit-done-cc --codex --global --uninstall
+npx get-shit-done-cc --kimi --global --uninstall
 
 # Local installs (current project)
 npx get-shit-done-cc --claude --local --uninstall