Files

T

YeonGyu-Kim 64825158a7 feat(agents): add Hephaestus - autonomous deep worker agent (#1287 )

* refactor(keyword-detector): split constants into domain-specific modules

* feat(shared): add requiresAnyModel and isAnyFallbackModelAvailable

* feat(config): add hephaestus to agent schemas

* feat(agents): add Hephaestus autonomous deep worker

* feat(cli): update model-fallback for hephaestus support

* feat(plugin): add hephaestus to config handler with ordering

* test(delegate-task): update tests for hephaestus agent

* docs: update AGENTS.md files for hephaestus

* docs: add hephaestus to READMEs

* chore: regenerate config schema

* fix(delegate-task): bypass requiresModel check when user provides explicit config

* docs(hephaestus): add 4-part context structure for explore/librarian prompts

* docs: fix review comments from cubic (non-breaking changes)

- Move Hephaestus from Primary Agents to Subagents (uses own fallback chain)
- Fix Hephaestus fallback chain documentation (claude-opus-4-5 → gemini-3-pro)
- Add settings.local.json to claude-code-hooks config sources
- Fix delegate_task parameters in ultrawork prompt (agent→subagent_type, background→run_in_background, add load_skills)
- Update line counts in AGENTS.md (index.ts: 788, manager.ts: 1440)

* docs: fix additional documentation inconsistencies from oracle review

- Fix delegate_task parameters in Background Agents example (docs/features.md)
- Fix Hephaestus fallback chain in root AGENTS.md to match model-requirements.ts

* docs: clarify Hephaestus has no fallback (requires gpt-5.2-codex only)

Hephaestus uses requiresModel constraint - it only activates when gpt-5.2-codex
is available. The fallback chain in code is unreachable, so documentation
should not mention fallbacks.

* fix(hephaestus): remove unreachable fallback chain entries

Hephaestus has requiresModel: gpt-5.2-codex which means the agent only
activates when that specific model is available. The fallback entries
(claude-opus-4-5, gemini-3-pro) were unreachable and misleading.

---------

Co-authored-by: justsisyphus <justsisyphus@users.noreply.github.com>

2026-02-01 19:26:57 +09:00

2.7 KiB

Raw Blame History

FEATURES KNOWLEDGE BASE

OVERVIEW

20 feature modules: background agents, skill MCPs, builtin skills/commands, Claude Code compatibility layer.

Feature Types: Task orchestration, Skill definitions, Command templates, Claude Code loaders, Supporting utilities

STRUCTURE

features/
├── background-agent/           # Task lifecycle (1418 lines)
│   ├── manager.ts              # Launch → poll → complete
│   └── concurrency.ts          # Per-provider limits
├── builtin-skills/             # Core skills (1729 lines)
│   └── skills.ts               # playwright, dev-browser, frontend-ui-ux, git-master, typescript-programmer
├── builtin-commands/           # ralph-loop, refactor, ulw-loop, init-deep, start-work, cancel-ralph, stop-continuation
├── claude-code-agent-loader/   # ~/.claude/agents/*.md
├── claude-code-command-loader/ # ~/.claude/commands/*.md
├── claude-code-mcp-loader/     # .mcp.json with ${VAR} expansion
├── claude-code-plugin-loader/  # installed_plugins.json
├── claude-code-session-state/  # Session persistence
├── opencode-skill-loader/      # Skills from 6 directories
├── context-injector/           # AGENTS.md/README.md injection
├── boulder-state/              # Todo state persistence
├── hook-message-injector/      # Message injection
├── task-toast-manager/         # Background task notifications
├── skill-mcp-manager/          # MCP client lifecycle (617 lines)
├── tmux-subagent/              # Tmux session management
├── mcp-oauth/                  # MCP OAuth handling
├── sisyphus-swarm/             # Swarm coordination
└── sisyphus-tasks/             # Task tracking

LOADER PRIORITY

Type	Priority (highest first)
Commands	`.opencode/command/` > `~/.config/opencode/command/` > `.claude/commands/`
Skills	`.opencode/skills/` > `~/.config/opencode/skills/` > `.claude/skills/`
MCPs	`.claude/.mcp.json` > `.mcp.json` > `~/.claude/.mcp.json`

BACKGROUND AGENT

Lifecycle: launch → poll (2s) → complete
Stability: 3 consecutive polls = idle
Concurrency: Per-provider/model limits via ConcurrencyManager
Cleanup: 30m TTL, 3m stale timeout
State: Per-session Maps, cleaned on session.deleted

SKILL MCP

Lazy: Clients created on first call
Transports: stdio, http (SSE/Streamable)
Lifecycle: 5m idle cleanup

ANTI-PATTERNS

Sequential delegation: Use delegate_task parallel
Trust self-reports: ALWAYS verify
Main thread blocks: No heavy I/O in loader init
Direct state mutation: Use managers for boulder/session state

2.7 KiB Raw Blame History