Development - #11
Merged
Merged
Conversation
…posal Record read-only inspection of the Command Code CLI v1.11.0 local data contract (~/.commandcode/projects/**/<session>.jsonl transcripts with per-message usage blocks) and propose a native collector behind the existing collector port.
…and detection Scope, acceptance criteria, test plan, and verified outcomes for the detection-only chunk. Command Code stays out of refresh targets and routing until the native collector is wired in a later chunk.
Introduce Command Code as an experimental Burnly source behind the collector port. Phase 1 wires identity and detection only: - SourceKey::CommandCode with stable storage key command-code and tray label Command Code. - CommandCodeCollector detection stub: scans ~/.commandcode/projects/** transcripts, distinguishes new-format (session version 3) from legacy pre-1.11 files, tolerates partial trailing lines, and fails closed on collect/describe until the native reader is wired. - Command Code rejected by ccusage routing and kept out of refresh targets (catalog stays 16). - Sanitized transcript fixtures with documented privacy constraints. - Product docs and README list Command Code as experimental.
…r and parser Scope, acceptance criteria, test plan, and verified outcomes for the read-only transcript foundation. The adapter remains fail-closed until the mapper is wired in a later chunk.
Phase 2 of the Command Code collector. Adds the read-only transcript foundation without changing collection behavior: - TranscriptReader scans projects/** for .jsonl session transcripts, skips checkpoint files, and tolerates partial trailing lines from live appends. - TranscriptParser decodes usage-only fields (never message content) from new-format session version 3 records, classifies legacy pre-1.11 files, and rejects negative tokens, token counts beyond i64, and invalid timestamps by skipping the line. - Edge-case fixtures for overflow, negative tokens, invalid timestamps, and missing usage fields.
Scope, acceptance criteria, test plan, and verified outcomes for the mapping layer. Cost provenance corrected to SourceReported per the Cline precedent.
Phase 3 of the Command Code collector. Adds the mapping layer without changing collection behavior: - map_daily aggregates parsed usage per local date with per-model breakdowns; map_sessions aggregates per (session id, model) with first/last activity. - Token fields map to input/output/cache-read/cache-write with checked overflow guards; canonical total is their sum. - costUsd converts to integer micros deterministically (round half-up) with SourceReported + Estimated provenance, matching the Cline convention. Zero cost with usage is Unavailable. - Records dedupe by (session id, message id) for re-read safety. - Engineering proposal cost decision updated accordingly.
Scope, acceptance criteria, test plan, and verified outcomes for wiring the collector into bootstrap, RoutedCollector, and the 18-target refresh catalog.
Phase 4 of the Command Code collector. The detection stub becomes a real Collector implementation: - collect/describe now run the transcript reader + mapper pipeline, with commandcode.collection_failed diagnostics carrying sanitized rowsFound counters; missing home returns Empty with a warning. - CommandCodeCollector is constructed in bootstrap/collectors.rs and routed in RoutedCollector (describe merges its profile). - Refresh target catalog grows 16 -> 18 (Command Code daily + session). - Dead-code allows removed from parser, reader, and mapper; parse transcript no longer returns Result since it cannot fail on a string.
Close the phase 4 wiring chunk and complete the collector roadmap with desktop runtime evidence: - Move phase 4 exec plan to completed/. - Capture runtime evidence at docs/runtime-evidence/2026-08-05-commandcode-runtime/: startup refresh succeeded, persisted 156,934,545 daily tokens for 2026-08-05 Asia/Jakarta (deepseek/deepseek-v4-flash, ~$11.29 provider estimate), 3 session rows, tray-summary query verified, and a privacy scan confirming zero conversation content in SQLite or logs. - Add active exec plan for phase 5 and cross-link the evidence from the engineering proposal.
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
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.
No description provided.