メニュー
Architecture, product-boundary, roadmap, and long-lived-rule consultation for harness-mem memory admission and review. Use when a candidate's truth depends on design judgment rather than only evidence lookup. Does not write memory.
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
| name | ask-memory-boundary |
| description | Architecture, product-boundary, roadmap, and long-lived-rule consultation for harness-mem memory admission and review. Use when a candidate's truth depends on design judgment rather than only evidence lookup. Does not write memory. |
| allowed-tools | ["Read","Grep","Glob"] |
Answer design-boundary questions for memory candidates.
This is the repo-local ask-me role for harness-mem. It helps decide how a
candidate should be scoped or worded when evidence alone is not enough:
Is this a product boundary decision?
Is this architecture truth, a workflow preference, or local task context?
Should this become a repo-wide rule, module documentation, or remain pending?
What are the tradeoffs and failure modes?
Use when grill-before-distill, /hm:review, or lookback raises a design
question:
govern_memory(action="suggest", kind="rule") or repo guidance.Return a concise consultation note:
recommendation: admit | narrow | defer | reject
classification: rule | memory_entry | handoff | relation | module_doc | local_note
scope: project | module | user-preference | task-local
reasoning:
- <tradeoff / boundary / risk>
safer_wording: <optional narrowed wording>
must_not_write_yet:
- <items that need proof or user/product decision>
govern_memory; this role only answers boundary questions.narrow or defer when the product boundary is uncertain.Run the harness-mem distill daily action when the user invokes hm-distill.
Run the harness-mem distill daily action when the user invokes hm-distill.
Evidence answerer for harness-mem memory admission and review. Use when grill-before-distill or review needs concrete support from packet evidence, observations, memory search, files, docs, tests, config, or command output. Does not write candidates or confirmed truth.
Standard memory admission mode for harness-mem (grill-me role). Runs automatically on distill — depth scales by risk, not a forced heavy loop every time. Three modes: deep interrogation, light checklist, lookback for stale truth. No writes. More than a review-stage helper: admission plus lookback.
Conversation autopilot for harness-mem. Use proactively when starting or resuming a task, when the user asks about previous decisions, when the user asks to remember a rule, when recent work becomes stable enough to preserve, or when a memory conflict appears.
Use harness-mem as a local-first memory runtime for the current project. Trigger when the user asks to remember prior work, resume a project, search old agent sessions, distill recent agent sessions, create durable project rules, or explain what the project currently knows.