Add Claude and orchestrated Codex skill usage import - #37
Merged
Conversation
Parse structured Claude Skill tool_use records with user and plugin source attribution, failure revisions, diagnostics, and symlink-safe artifact deduplication. Harden source identity aggregation, bounded partition reads, coverage messaging, and Windows artifact containment across core and TUI. Closes #27 Refs #33, #34, #35
Support current custom_tool_call exec envelopes with bounded Acorn-based static evidence extraction while retaining legacy imports. Document provider-specific cursor migration, evidence status, diagnostics, and safety limits.
Map @skillpack/core to its workspace source for type checking so pnpm check:ci does not depend on a pre-existing Core declaration build.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Skilltool calls from project and subagent JSONL artifacts, with user/plugin source attribution, evidence-based statuses, append-only revisions, diagnostics, and symlink-safe artifact deduplication.custom_tool_call/execenvelopes through a bounded Acorn-based static evaluator while preserving historicalexec_commandimports.Why
Skill Usage previously covered only the historical Codex execution envelope. Claude invocations were not imported, and current Codex orchestration wrapped skill reads in JavaScript that the importer could not recognize. This change adds structured Claude support and safely extracts only statically provable Codex skill-read evidence without executing session code.
User impact
Users who explicitly enable Skill Usage import can now see aggregate Claude usage alongside Codex usage. Current orchestrated Codex sessions are counted again, source aliases and plugin cache versions aggregate more accurately, and ambiguous evidence fails closed or surfaces bounded diagnostics. Skillpack continues to retain only minimal invocation metadata rather than prompts, responses, tool arguments, commands, or transcripts.
Validation
pnpm checkpnpm test— 194 tests passedpnpm build:ci— Core/TUI packaging and npm artifact verification passedCloses #27
Refs #33, #34, #35