feat: Import claude.ai data#8
Merged
Merged
Conversation
added 8 commits
February 16, 2025 10:33
phernandez
added a commit
that referenced
this pull request
May 29, 2026
Resolves the 8 self-review findings on PR #865: session-start.sh: - Wire recallTimeframe into a new "Recent sessions" query (--after_date) — it was parsed but unused after the Phase 4 rewrite. The brief now surfaces recent session checkpoints (the resume cursor), which is the one query the recency window legitimately applies to. (#1) - Guard secondaryProjects/teamProjects JSON types — a string value was iterated character-by-character into bogus per-char queries. (#3) - A configured-but-unreachable/misnamed primaryProject now emits a one-line "couldn't read" signal instead of a silent blank brief. (#7) pre-compact.sh: - Filter transcript turns by the `isMeta` and `toolUseResult` flags instead of a `text.startswith("<")` heuristic. This stops dropping legitimate user messages that start with "<" (#4) and stops capturing tool-result/meta frames as human turns (#8) — verified against a real transcript (25 human turns cleanly separated from 8 meta + 288 tool-result frames). - Title now uses second precision so rapid same-minute compactions don't collide and silently drop/overwrite a checkpoint; removed the unused stamp/slug. (#2) - Empty-checkpoint guard now requires at least one real user turn, not just any turn, so an assistant-only transcript can't produce a dangling-title note. (#5) validate_skills.py: - parse_frontmatter now skips indented lines, capturing only top-level keys, so a schema note's nested schema:/settings: children can't overwrite a top-level type/entity via last-write-wins. Documented its single-line-only limitation. (#6) All verified end-to-end; `just package-check-claude-code` and `package-check-skills` pass; ruff clean. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Signed-off-by: phernandez <paul@basicmachines.co>
phernandez
added a commit
that referenced
this pull request
May 30, 2026
Resolves the 8 self-review findings on PR #865: session-start.sh: - Wire recallTimeframe into a new "Recent sessions" query (--after_date) — it was parsed but unused after the Phase 4 rewrite. The brief now surfaces recent session checkpoints (the resume cursor), which is the one query the recency window legitimately applies to. (#1) - Guard secondaryProjects/teamProjects JSON types — a string value was iterated character-by-character into bogus per-char queries. (#3) - A configured-but-unreachable/misnamed primaryProject now emits a one-line "couldn't read" signal instead of a silent blank brief. (#7) pre-compact.sh: - Filter transcript turns by the `isMeta` and `toolUseResult` flags instead of a `text.startswith("<")` heuristic. This stops dropping legitimate user messages that start with "<" (#4) and stops capturing tool-result/meta frames as human turns (#8) — verified against a real transcript (25 human turns cleanly separated from 8 meta + 288 tool-result frames). - Title now uses second precision so rapid same-minute compactions don't collide and silently drop/overwrite a checkpoint; removed the unused stamp/slug. (#2) - Empty-checkpoint guard now requires at least one real user turn, not just any turn, so an assistant-only transcript can't produce a dangling-title note. (#5) validate_skills.py: - parse_frontmatter now skips indented lines, capturing only top-level keys, so a schema note's nested schema:/settings: children can't overwrite a top-level type/entity via last-write-wins. Documented its single-line-only limitation. (#6) All verified end-to-end; `just package-check-claude-code` and `package-check-skills` pass; ruff clean. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Signed-off-by: phernandez <paul@basicmachines.co>
phernandez
added a commit
that referenced
this pull request
May 31, 2026
Resolves the 8 self-review findings on PR #865: session-start.sh: - Wire recallTimeframe into a new "Recent sessions" query (--after_date) — it was parsed but unused after the Phase 4 rewrite. The brief now surfaces recent session checkpoints (the resume cursor), which is the one query the recency window legitimately applies to. (#1) - Guard secondaryProjects/teamProjects JSON types — a string value was iterated character-by-character into bogus per-char queries. (#3) - A configured-but-unreachable/misnamed primaryProject now emits a one-line "couldn't read" signal instead of a silent blank brief. (#7) pre-compact.sh: - Filter transcript turns by the `isMeta` and `toolUseResult` flags instead of a `text.startswith("<")` heuristic. This stops dropping legitimate user messages that start with "<" (#4) and stops capturing tool-result/meta frames as human turns (#8) — verified against a real transcript (25 human turns cleanly separated from 8 meta + 288 tool-result frames). - Title now uses second precision so rapid same-minute compactions don't collide and silently drop/overwrite a checkpoint; removed the unused stamp/slug. (#2) - Empty-checkpoint guard now requires at least one real user turn, not just any turn, so an assistant-only transcript can't produce a dangling-title note. (#5) validate_skills.py: - parse_frontmatter now skips indented lines, capturing only top-level keys, so a schema note's nested schema:/settings: children can't overwrite a top-level type/entity via last-write-wins. Documented its single-line-only limitation. (#6) All verified end-to-end; `just package-check-claude-code` and `package-check-skills` pass; ruff clean. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Signed-off-by: phernandez <paul@basicmachines.co>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Import Claude.ai conversation and project data to basic-memory Markdown format.