菜单
recursive investigation orchestration using harness forge. use when the task is hard, ambiguous, cross-module, or likely to benefit from Typed RLM, bounded subcalls, durable iterations, and replayable recursive artifacts instead of chat-only reasoning.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Auto-discoverable wrapper for `.hforge/library/skills/bug-investigation/SKILL.md`.
Auto-discoverable wrapper for `.hforge/library/skills/double-diamond-feature/SKILL.md`.
Use for defects, regressions, outages, flaky tests, and unexpected behavior. Guides the agent through triage, containment, reproduction, hypotheses, root-cause evidence, fix verification, and recurrence prevention. Do not use as the primary workflow for new features; use double-diamond-feature instead.
Use for meaningful feature development before coding. Guides the agent through Discover, Define, Develop, and Deliver with evidence, options, validation, and rollback notes. Do not use for bugs, regressions, outages, or flaky tests; use bug-investigation instead.
Auto-discoverable wrapper for `.hforge/library/skills/complex-task-protocol/SKILL.md`.
Automatic operating protocol for complex agent tasks. Use when work is multi-step, risky, cross-module, orchestration-heavy, or likely to benefit from bounded sidecar help, verification discipline, recovery checkpoints, and durable learning capture.
| name | hforge-recursive-investigate |
| description | recursive investigation orchestration using harness forge. use when the task is hard, ambiguous, cross-module, or likely to benefit from Typed RLM, bounded subcalls, durable iterations, and replayable recursive artifacts instead of chat-only reasoning. |
Do not use this for straightforward single-file edits where ordinary repo inspection is enough.
AGENTS.md.hforge/agent-manifest.json.hforge/generated/agent-command-catalog.json.hforge/runtime/index.json.hforge/runtime/recursive/language-capabilities.json.hforge/runtime/recursive/runtime-inventory.json.hforge/runtime/recursive/escalation-heuristics.json.hforge/runtime/recursive/sessions/commands/hforge-recursive-investigate.mdhforge status --root . --jsonhforge bootstrap --root . --yeshforge recursive capabilities --root . --jsonhforge recursive runtimes --root . --jsonhforge recursive provision-runtime <python|powershell> --root . --json only when a workspace-managed alias is helpfulhforge recursive plan "<objective>" --task-id <taskId> --root . --jsonhforge recursive execute with a typed bundle before using freeform structured-analysis snippetshforge recursive run only for one bounded fallback step when Typed RLM is not the right fititerations, subcalls, cells, promotions, meta-ops, score, and replay to summarize the work from durable artifactsA good recursive-investigate result should include: