メニュー
Auto-discoverable wrapper for the Harness Forge recursive-investigate workflow. Use when a task is ambiguous, cross-module, investigation-heavy, or likely to benefit from Typed RLM, bounded subcalls, and durable recursive artifacts.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
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 | Auto-discoverable wrapper for the Harness Forge recursive-investigate workflow. Use when a task is ambiguous, cross-module, investigation-heavy, or likely to benefit from Typed RLM, bounded subcalls, and durable recursive artifacts. |
| compatibility | Works best in repositories already initialized with Harness Forge and should remain honest about translated support on partial targets. |
| metadata | {"author":"harness-forge","source":"skills/hforge-recursive-investigate/SKILL.md"} |
$ARGUMENTS
You must consider the user input before proceeding.
Use Harness Forge recursive mode as an explicit investigation operating layer for hard tasks, preferring Typed RLM first and only falling back to bounded structured analysis when necessary.
/hforge-recursive-investigate as the investigation objective or task focus..hforge/agent-manifest.json, .hforge/runtime/index.json, .hforge/runtime/recursive/language-capabilities.json, .hforge/runtime/recursive/runtime-inventory.json, and .hforge/runtime/recursive/escalation-heuristics.json first when present.hforge recursive capabilities --root . --json before claiming recursive support for the active language or target.hforge recursive runtimes --root . --json before assuming Python or PowerShell code-cell availability.hforge recursive provision-runtime <python|powershell> --root . --json only when a workspace-managed runtime alias is genuinely useful.hforge recursive plan "<objective>" --task-id <taskId> --root . --json or hforge recursive inspect <sessionId> --root . --json.hforge recursive execute <sessionId> --file <bundle.json> --root . --json or --stdin.hforge recursive run <sessionId> --file <snippet> or --stdin only when one bounded structured-analysis step is the right tool.hforge recursive iterations, subcalls, cells, promotions, meta-ops, score, and replay.