feat: add targeted skill improvements and Copilot CLI install docs#7
feat: add targeted skill improvements and Copilot CLI install docs#7tyler555g wants to merge 6 commits intobasicmachines-co:mainfrom
Conversation
- memory-tasks: formalize Resume/Pause protocols with SEARCH→READ→VALIDATE→CONTINUE sequence - memory-tasks: add "Why Both Frontmatter and Observations?" subsection - memory-defrag: add resolved error traces as explicit cleanup target with example - memory-schema: add dual-location rationale (why fields in both frontmatter and observations) - memory-metadata-search: add consistency-enables-filtering guideline - README: add GitHub Copilot CLI to Compatible Agents and install examples Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
There was a problem hiding this comment.
Pull request overview
This PR is a documentation-focused follow-up that tightens several Basic Memory skill guides (tasks, schema, metadata search, defrag) and updates the README with GitHub Copilot CLI installation examples.
Changes:
- Expanded/structured rationale for “frontmatter vs observations” and added more explicit resume/pause protocols in
memory-tasksandmemory-schema. - Added “resolved error traces” as a first-class defrag target plus checklist/log updates in
memory-defrag. - Added a consistency guideline for metadata filtering in
memory-metadata-search, and added Copilot CLI install examples toREADME.md.
Reviewed changes
Copilot reviewed 5 out of 5 changed files in this pull request and generated 3 comments.
Show a summary per file
| File | Description |
|---|---|
| memory-tasks/SKILL.md | Adds structured Resume/Pause protocols and clarifies dual-location fields (frontmatter + observations). |
| memory-schema/SKILL.md | Replaces a single warning paragraph with a structured explanation of why fields may need to exist in two places. |
| memory-metadata-search/SKILL.md | Adds a guideline about consistent metadata values to keep filters effective. |
| memory-defrag/SKILL.md | Adds “resolved error traces” as a cleanup category plus checklist/log template updates. |
| README.md | Adds GitHub Copilot CLI install command and lists Copilot CLI among compatible agents. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
- Clarify note_type is normalized to lowercase, not case-sensitive (memory-tasks) - Use --agent flag consistently instead of -a shorthand (README) - Update skill count from 9 to 10 to include memory-literary-analysis (README) Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 5 out of 5 changed files in this pull request and generated 3 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
- Clarify current_step semantics in Resume Protocol (memory-tasks) - Nuance frontmatter-only field validation behavior (memory-schema) - Use archive language instead of remove for completed tasks (memory-defrag) Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 5 out of 5 changed files in this pull request and generated 4 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
…y context engineering principles - Fix type casing: Task -> task for consistency with normalization (memory-tasks) - Fix stale section reference: Pre-compaction flush -> Pause protocol (memory-tasks) - Clarify schema-required fields must exist as observations (memory-tasks) - Update skill count from 9 to 10 in README tip - Add context engineering connections: context rot/distraction (memory-defrag), Select principle (memory-metadata-search), task note density (memory-tasks) - CE references link to tyler555g/best-practices context-engineering.md Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 5 out of 5 changed files in this pull request and generated 2 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Replace ellipsis placeholders in edit_note find_replace examples with exact text that matches the corresponding before/template content shown earlier in each file. Ellipsis strings would fail to match real note content at runtime. - memory-tasks: match the Context header line from the Creating a Task template - memory-defrag: match the three error trace lines shown in the Before block Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 5 out of 5 changed files in this pull request and generated 2 comments.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
- memory-tasks: find_text now covers the full Context template block
including all four bullet lines, so replacement doesn't leave
the original scaffold behind
- memory-defrag: fix grammar in orphan files table row
('never referenced' → 'that are never referenced')
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
There was a problem hiding this comment.
Pull request overview
Copilot reviewed 5 out of 5 changed files in this pull request and generated 1 comment.
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
| - **Omit `query` for filter-only searches.** `search_notes(metadata_filters={"status": "active"})` works without a text query. | ||
| - **Dot notation for nesting.** Access nested YAML structures with dots: `{"schema.version": "2"}` queries the `version` key inside a `schema` object. | ||
| - **Tags shortcut is convenient but limited.** `tags` and `status` are sugar for common fields. For anything else, use `metadata_filters` directly. | ||
| - **Consistency enables filtering.** Metadata filters require consistent values across notes. If you use `status: active` sometimes and `status: in-progress` other times for the same concept, filters break. Use schemas to enforce consistency. |
There was a problem hiding this comment.
The guideline says to “Use schemas to enforce consistency,” but schema_validate explicitly ignores frontmatter-only fields (it only checks observation categories). As written, this implies schemas will enforce metadata filter values even when they’re only in frontmatter, which conflicts with the behavior described in memory-schema/SKILL.md. Consider rephrasing to clarify that schemas enforce consistency only for fields present as observations (and that metadata-filtered fields should be kept in sync across frontmatter + observations if you want both filtering and validation).
| - **Consistency enables filtering.** Metadata filters require consistent values across notes. If you use `status: active` sometimes and `status: in-progress` other times for the same concept, filters break. Use schemas to enforce consistency. | |
| - **Consistency enables filtering.** Metadata filters require consistent values across notes. If you use `status: active` sometimes and `status: in-progress` other times for the same concept, filters break. Schemas help enforce consistency only for fields that are also recorded as observations and validated by `schema_validate`; frontmatter-only fields are still searchable, but if you want both filtering and validation, keep the frontmatter value and corresponding observation in sync. |
Summary
Focused follow-up to #6, containing only the specific content improvements identified as useful during review. No infrastructure changes, no 12-factor branding, no README restructuring.
Changes
memory-tasks/SKILL.md
current_stepagainst checkboxesnote_types→note_typetypo in Key Principlesmemory-defrag/SKILL.md
memory-schema/SKILL.md
metadata_filters) and observations (forschema_validate)memory-metadata-search/SKILL.md
README.md
npx skills add -a github-copilotinstall commandContext
These changes were identified during review of #6. All match the existing skill voice (declarative, terse, no epigraphs).