jedarden/claude-print
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Update bead tracking for Phase 6 completion (bf-64s)

Browse source 
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
jedarden 2026-06-10 00:17:20 -04:00
parent ab1e9d6d64
commit ed7790f6e2
1 changed files with 7 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

14
.beads/issues.jsonl
View file

@ -1,7 +1,7 @@
{"id":"bf-10t","title":"Phase 10: Tests (~500 LOC)","description":"Entry: Phase 8 complete (can run in parallel with Phase 9).\n\nPhase 10 completes the test suite by adding tests NOT already written as part of Phases 2-9 completion criteria. Each prior phase's completion criterion already specifies and runs its own targeted integration tests.\n\nPhase 10 adds the remaining cross-phase and corner-case tests:\n- Version-resilience suite: feed transcript JSONL with extra/unknown fields across all JSONL event types; confirm no panic, correct output\n- Hook inheritance suite: verify user hooks in ~/.claude/settings.json fire alongside the relay hook (OQ-2 resolution validated end-to-end)\n- All MEDIUM/LOW mock scenarios not covered by earlier phases (see Testing section of plan.md for full list)\n- Conformance harness: run claude-print against a real claude binary in a sandboxed invocation and compare output format to claude -p reference\n\nComplete when:\n- cargo test passes with zero failures (all unit + integration tests)\n\nReference: docs/plan/plan.md \u00a7 Phase 10","design":"","acceptance_criteria":"","notes":"","status":"open","priority":1,"issue_type":"task","created_at":"2026-06-10T03:54:17.744030380Z","updated_at":"2026-06-10T03:54:17.744030380Z","source_repo":".","compaction_level":0,"dependencies":[{"issue_id":"bf-10t","depends_on_id":"bf-2f1","type":"blocks","created_at":"2026-06-10T03:54:31.710642868Z","created_by":"cli","thread_id":""}]}
{"id":"bf-2f1","title":"Phase 8: Emitter (~120 LOC)","description":"Entry: Phase 7 complete.\n\nImplementation: src/emitter.rs\n- Output format routing: text (plain string to stdout), json (JSON object with result/cost/session_id/claude_version), stream-json (forward raw JSONL lines as written)\n- claude_version: read from child process version string captured at startup\n- Error result objects: exit code mapping (0=success, 1=error, 2=no-response, 3=timeout, 4=input-error)\n- stream-json: reader thread consuming JSONL transcript file via mpsc channel, forward each line to stdout as written\n\nComplete when:\n- All emitter unit tests pass (tests/emitter.rs)\n- AS-1 (text output format) passes: response text emitted to stdout, exit 0\n- AS-2 (json output format) passes: valid JSON object on stdout with result field\n- stream-json output parses as valid JSONL (each line is valid JSON)\n\nReference: docs/plan/plan.md \u00a7 Phase 8","design":"","acceptance_criteria":"","notes":"","status":"open","priority":1,"issue_type":"task","created_at":"2026-06-10T03:53:59.176338208Z","updated_at":"2026-06-10T03:53:59.176338208Z","source_repo":".","compaction_level":0,"dependencies":[{"issue_id":"bf-2f1","depends_on_id":"bf-64k","type":"blocks","created_at":"2026-06-10T03:54:31.701710128Z","created_by":"cli","thread_id":""}]}
{"id":"bf-42j","title":"Phase 9: NEEDLE Integration (~50 LOC + config)","description":"Entry: Phase 8 complete.\n\nDeliverables:\n1. claude-print.yaml \u2014 NEEDLE agent config for dispatching beads via claude-print instead of claude -p\n - input_method: stdin, output_transform: needle-transform-claude\n - invoke_template: claude-print --output-format json --model ${MODEL}\n2. install.sh \u2014 download release binary from GitHub, install to ~/.local/bin/claude-print, verify --check\n3. claude-print-ci WorkflowTemplate in jedarden/declarative-config (k8s/iad-ci/argo-workflows/)\n - verify step only (build-musl + github-release steps added in Phase 11)\n - delegates to rust-verify WorkflowTemplate\n4. --check subcommand in src/main.rs or src/check.rs\n - openpty probe: confirm openpty syscall succeeds\n - mkfifo probe: confirm mkfifo in /home/coding/.tmp succeeds\n - optional mock_claude PTY round-trip (if mock_claude binary present in PATH)\n - exits 0 on success, prints diagnostic table\n\nComplete when:\n- bash -n install.sh passes (syntactically valid)\n- Manually copying locally-built binary to ~/.local/bin/claude-print and running claude-print --check succeeds\n- NEEDLE dispatches a test bead using claude-print.yaml; AS-3 passes\n- README flags table matches claude-print --help output exactly (verified manually)\n\nReference: docs/plan/plan.md \u00a7 Phase 9","design":"","acceptance_criteria":"","notes":"","status":"open","priority":1,"issue_type":"task","created_at":"2026-06-10T03:54:09.318382701Z","updated_at":"2026-06-10T03:54:09.318382701Z","source_repo":".","compaction_level":0,"dependencies":[{"issue_id":"bf-42j","depends_on_id":"bf-2f1","type":"blocks","created_at":"2026-06-10T03:54:31.706113490Z","created_by":"cli","thread_id":""}]}
{"id":"bf-4no","title":"Phase 11: CI (~YAML only)","description":"Entry: Phase 10 complete (and Phase 9 complete for install.sh e2e test).\n\nDeliverables in jedarden/declarative-config (k8s/iad-ci/argo-workflows/claude-print-ci.yaml):\n- Update claude-print-ci WorkflowTemplate stub (from Phase 9) with full steps:\n 1. verify \u2014 delegates to rust-verify WorkflowTemplate (fmt + clippy + test + cargo audit)\n 2. build-musl \u2014 cross-compile x86_64-unknown-linux-musl release binary\n 3. build-mock-claude-musl \u2014 build test-fixtures/mock-claude/ as musl binary\n 4. github-release \u2014 upload claude-print + mock_claude binaries + last-claude-version.txt artifact to GitHub Release\n- Confirm cargo audit runs (either via rust-verify or as explicit step between verify and build-musl)\n- install.sh end-to-end download test: download release artifact from GitHub Release URL, verify install.sh exits 0 and claude-print --check passes\n\nComplete when:\n- CI run on main branch produces release binary at expected GitHub Release URL\n- last-claude-version.txt artifact present in release\n- Binary passes claude-print --check (credential-free) via install.sh\n- install.sh end-to-end download test passes (deferred from Phase 9)\n- AS-1 verified manually before pushing release tag\n\nReference: docs/plan/plan.md \u00a7 Phase 11","design":"","acceptance_criteria":"","notes":"","status":"open","priority":1,"issue_type":"task","created_at":"2026-06-10T03:54:27.444014247Z","updated_at":"2026-06-10T03:54:27.444014247Z","source_repo":".","compaction_level":0,"dependencies":[{"issue_id":"bf-4no","depends_on_id":"bf-10t","type":"blocks","created_at":"2026-06-10T03:54:31.717358160Z","created_by":"cli","thread_id":""},{"issue_id":"bf-4no","depends_on_id":"bf-42j","type":"blocks","created_at":"2026-06-10T03:54:31.725797267Z","created_by":"cli","thread_id":""}]}
{"id":"bf-5bl","title":"Starvation alert: beads invisible to worker","description":"## Starvation Alert\n\nOpen beads exist but Pluck found none \u2014 possible configuration error.\n\n**Workspace:** default\n**Total beads:** 6\n**Open:** 5\n**In-progress:** 1\n**Claimed by:** claude-glm-glm47-alpha\n\nCheck exclude_labels, workspace path, and filter configuration.","design":"","acceptance_criteria":"","notes":"","status":"closed","priority":2,"issue_type":"task","assignee":null,"created_at":"2026-06-10T03:57:06.245148475Z","updated_at":"2026-06-10T04:30:00.000000000Z","source_repo":".","compaction_level":0,"labels":["starvation-alert"],"closed_at":"2026-06-10T04:30:00.000000000Z"}
{"id":"bf-64k","title":"Phase 7: Transcript Reader (~180 LOC)","description":"Entry: Phase 6 complete. PO-5 acknowledged: retry loop (40\u00d750ms) is the mitigation for Stop-before-JSONL races. Verify retry timing by running test_transcript_race with MOCK_DELAY_JSONL=100 and confirming exit 0.\n\nImplementation: src/transcript.rs\n- JSONL parse with lenient serde (unknown fields tolerated, unknown event types skipped)\n- message.id dedup + usage-fingerprint fallback dedup for events without message.id\n- Text extraction from assistant ContentBlock array (text type only)\n- 40\u00d750ms retry loop with Stop-payload fallback to last_assistant_message after exhausted\n- Path derivation: strip leading /, replace / with -, append session_id from Stop payload\n\nComplete when:\n- All transcript unit tests pass (tests/transcript.rs)\n- test_streaming_dedup_40_retries passes\n- AS-6 (race scenario: Stop fires before JSONL flush) passes with MOCK_DELAY_JSONL=100\n\nReference: docs/plan/plan.md \u00a7 Phase 7","design":"","acceptance_criteria":"","notes":"","status":"open","priority":1,"issue_type":"task","created_at":"2026-06-10T03:53:52.452812786Z","updated_at":"2026-06-10T03:53:52.452812786Z","source_repo":".","compaction_level":0,"dependencies":[{"issue_id":"bf-64k","depends_on_id":"bf-64s","type":"blocks","created_at":"2026-06-10T03:54:31.695602796Z","created_by":"cli","thread_id":""}]}
{"id":"bf-64s","title":"Phase 6: Stop Poller (~80 LOC)","description":"Entry: Phase 5 complete. OQ-2 must be resolved (verify --setting-sources= suppresses standard sources; see PO-2 for fallback). OQ-4 (FIFO open race) validated by test.\n\nImplementation: poller.rs (or extend event_loop.rs)\n- Open FIFO read-end O_NONBLOCK, integrate into poll() loop\n- Parse Stop hook JSON payload (session_id, transcript_path, last_assistant_message)\n- Derive transcript path from session_id + cwd slug if transcript_path absent\n- Signal event loop exit via channel/flag\n\nComplete when:\n- Integration test test_stop_hook_fires passes (mock_claude emits Stop, FIFO received, exit 0)\n- test_missing_transcript_path_derived passes (Stop without transcript_path \u2192 path derived from session_id)\n\nReference: docs/plan/plan.md \u00a7 Phase 6","design":"","acceptance_criteria":"","notes":"","status":"in_progress","priority":1,"issue_type":"task","assignee":"claude-glm-glm47-alpha","created_at":"2026-06-10T03:53:44.914912586Z","updated_at":"2026-06-10T03:54:27.001527172Z","source_repo":".","compaction_level":0}
{"id":"bf-10t","title":"Phase 10: Tests (~500 LOC)","description":"Entry: Phase 8 complete (can run in parallel with Phase 9).\n\nPhase 10 completes the test suite by adding tests NOT already written as part of Phases 2-9 completion criteria. Each prior phase's completion criterion already specifies and runs its own targeted integration tests.\n\nPhase 10 adds the remaining cross-phase and corner-case tests:\n- Version-resilience suite: feed transcript JSONL with extra/unknown fields across all JSONL event types; confirm no panic, correct output\n- Hook inheritance suite: verify user hooks in ~/.claude/settings.json fire alongside the relay hook (OQ-2 resolution validated end-to-end)\n- All MEDIUM/LOW mock scenarios not covered by earlier phases (see Testing section of plan.md for full list)\n- Conformance harness: run claude-print against a real claude binary in a sandboxed invocation and compare output format to claude -p reference\n\nComplete when:\n- cargo test passes with zero failures (all unit + integration tests)\n\nReference: docs/plan/plan.md § Phase 10","design":"","acceptance_criteria":"","notes":"","status":"open","priority":1,"issue_type":"task","created_at":"2026-06-10T03:54:17.744030380Z","updated_at":"2026-06-10T03:54:17.744030380Z","source_repo":".","compaction_level":0,"dependencies":[{"issue_id":"bf-10t","depends_on_id":"bf-2f1","type":"blocks","created_at":"2026-06-10T03:54:31.710642868Z","created_by":"cli","thread_id":""}]}
{"id":"bf-2f1","title":"Phase 8: Emitter (~120 LOC)","description":"Entry: Phase 7 complete.\n\nImplementation: src/emitter.rs\n- Output format routing: text (plain string to stdout), json (JSON object with result/cost/session_id/claude_version), stream-json (forward raw JSONL lines as written)\n- claude_version: read from child process version string captured at startup\n- Error result objects: exit code mapping (0=success, 1=error, 2=no-response, 3=timeout, 4=input-error)\n- stream-json: reader thread consuming JSONL transcript file via mpsc channel, forward each line to stdout as written\n\nComplete when:\n- All emitter unit tests pass (tests/emitter.rs)\n- AS-1 (text output format) passes: response text emitted to stdout, exit 0\n- AS-2 (json output format) passes: valid JSON object on stdout with result field\n- stream-json output parses as valid JSONL (each line is valid JSON)\n\nReference: docs/plan/plan.md § Phase 8","design":"","acceptance_criteria":"","notes":"","status":"open","priority":1,"issue_type":"task","created_at":"2026-06-10T03:53:59.176338208Z","updated_at":"2026-06-10T03:53:59.176338208Z","source_repo":".","compaction_level":0,"dependencies":[{"issue_id":"bf-2f1","depends_on_id":"bf-64k","type":"blocks","created_at":"2026-06-10T03:54:31.701710128Z","created_by":"cli","thread_id":""}]}
{"id":"bf-42j","title":"Phase 9: NEEDLE Integration (~50 LOC + config)","description":"Entry: Phase 8 complete.\n\nDeliverables:\n1. claude-print.yaml — NEEDLE agent config for dispatching beads via claude-print instead of claude -p\n - input_method: stdin, output_transform: needle-transform-claude\n - invoke_template: claude-print --output-format json --model ${MODEL}\n2. install.sh — download release binary from GitHub, install to ~/.local/bin/claude-print, verify --check\n3. claude-print-ci WorkflowTemplate in jedarden/declarative-config (k8s/iad-ci/argo-workflows/)\n - verify step only (build-musl + github-release steps added in Phase 11)\n - delegates to rust-verify WorkflowTemplate\n4. --check subcommand in src/main.rs or src/check.rs\n - openpty probe: confirm openpty syscall succeeds\n - mkfifo probe: confirm mkfifo in /home/coding/.tmp succeeds\n - optional mock_claude PTY round-trip (if mock_claude binary present in PATH)\n - exits 0 on success, prints diagnostic table\n\nComplete when:\n- bash -n install.sh passes (syntactically valid)\n- Manually copying locally-built binary to ~/.local/bin/claude-print and running claude-print --check succeeds\n- NEEDLE dispatches a test bead using claude-print.yaml; AS-3 passes\n- README flags table matches claude-print --help output exactly (verified manually)\n\nReference: docs/plan/plan.md § Phase 9","design":"","acceptance_criteria":"","notes":"","status":"open","priority":1,"issue_type":"task","created_at":"2026-06-10T03:54:09.318382701Z","updated_at":"2026-06-10T03:54:09.318382701Z","source_repo":".","compaction_level":0,"dependencies":[{"issue_id":"bf-42j","depends_on_id":"bf-2f1","type":"blocks","created_at":"2026-06-10T03:54:31.706113490Z","created_by":"cli","thread_id":""}]}
{"id":"bf-4no","title":"Phase 11: CI (~YAML only)","description":"Entry: Phase 10 complete (and Phase 9 complete for install.sh e2e test).\n\nDeliverables in jedarden/declarative-config (k8s/iad-ci/argo-workflows/claude-print-ci.yaml):\n- Update claude-print-ci WorkflowTemplate stub (from Phase 9) with full steps:\n 1. verify — delegates to rust-verify WorkflowTemplate (fmt + clippy + test + cargo audit)\n 2. build-musl — cross-compile x86_64-unknown-linux-musl release binary\n 3. build-mock-claude-musl — build test-fixtures/mock-claude/ as musl binary\n 4. github-release — upload claude-print + mock_claude binaries + last-claude-version.txt artifact to GitHub Release\n- Confirm cargo audit runs (either via rust-verify or as explicit step between verify and build-musl)\n- install.sh end-to-end download test: download release artifact from GitHub Release URL, verify install.sh exits 0 and claude-print --check passes\n\nComplete when:\n- CI run on main branch produces release binary at expected GitHub Release URL\n- last-claude-version.txt artifact present in release\n- Binary passes claude-print --check (credential-free) via install.sh\n- install.sh end-to-end download test passes (deferred from Phase 9)\n- AS-1 verified manually before pushing release tag\n\nReference: docs/plan/plan.md § Phase 11","design":"","acceptance_criteria":"","notes":"","status":"open","priority":1,"issue_type":"task","created_at":"2026-06-10T03:54:27.444014247Z","updated_at":"2026-06-10T03:54:27.444014247Z","source_repo":".","compaction_level":0,"dependencies":[{"issue_id":"bf-4no","depends_on_id":"bf-10t","type":"blocks","created_at":"2026-06-10T03:54:31.717358160Z","created_by":"cli","thread_id":""},{"issue_id":"bf-4no","depends_on_id":"bf-42j","type":"blocks","created_at":"2026-06-10T03:54:31.725797267Z","created_by":"cli","thread_id":""}]}
{"id":"bf-5bl","title":"Starvation alert: beads invisible to worker","description":"## Starvation Alert\n\nOpen beads exist but Pluck found none — possible configuration error.\n\n**Workspace:** default\n**Total beads:** 6\n**Open:** 5\n**In-progress:** 1\n**Claimed by:** claude-glm-glm47-alpha\n\nCheck exclude_labels, workspace path, and filter configuration.","design":"","acceptance_criteria":"","notes":"","status":"closed","priority":2,"issue_type":"task","created_at":"2026-06-10T03:57:06.245148475Z","updated_at":"2026-06-10T04:30:00Z","closed_at":"2026-06-10T04:30:00Z","source_repo":".","compaction_level":0,"labels":["starvation-alert"]}
{"id":"bf-64k","title":"Phase 7: Transcript Reader (~180 LOC)","description":"Entry: Phase 6 complete. PO-5 acknowledged: retry loop (40×50ms) is the mitigation for Stop-before-JSONL races. Verify retry timing by running test_transcript_race with MOCK_DELAY_JSONL=100 and confirming exit 0.\n\nImplementation: src/transcript.rs\n- JSONL parse with lenient serde (unknown fields tolerated, unknown event types skipped)\n- message.id dedup + usage-fingerprint fallback dedup for events without message.id\n- Text extraction from assistant ContentBlock array (text type only)\n- 40×50ms retry loop with Stop-payload fallback to last_assistant_message after exhausted\n- Path derivation: strip leading /, replace / with -, append session_id from Stop payload\n\nComplete when:\n- All transcript unit tests pass (tests/transcript.rs)\n- test_streaming_dedup_40_retries passes\n- AS-6 (race scenario: Stop fires before JSONL flush) passes with MOCK_DELAY_JSONL=100\n\nReference: docs/plan/plan.md § Phase 7","design":"","acceptance_criteria":"","notes":"","status":"open","priority":1,"issue_type":"task","created_at":"2026-06-10T03:53:52.452812786Z","updated_at":"2026-06-10T03:53:52.452812786Z","source_repo":".","compaction_level":0,"dependencies":[{"issue_id":"bf-64k","depends_on_id":"bf-64s","type":"blocks","created_at":"2026-06-10T03:54:31.695602796Z","created_by":"cli","thread_id":""}]}
{"id":"bf-64s","title":"Phase 6: Stop Poller (~80 LOC)","description":"Entry: Phase 5 complete. OQ-2 must be resolved (verify --setting-sources= suppresses standard sources; see PO-2 for fallback). OQ-4 (FIFO open race) validated by test.\n\nImplementation: poller.rs (or extend event_loop.rs)\n- Open FIFO read-end O_NONBLOCK, integrate into poll() loop\n- Parse Stop hook JSON payload (session_id, transcript_path, last_assistant_message)\n- Derive transcript path from session_id + cwd slug if transcript_path absent\n- Signal event loop exit via channel/flag\n\nComplete when:\n- Integration test test_stop_hook_fires passes (mock_claude emits Stop, FIFO received, exit 0)\n- test_missing_transcript_path_derived passes (Stop without transcript_path → path derived from session_id)\n\nReference: docs/plan/plan.md § Phase 6","design":"","acceptance_criteria":"","notes":"","status":"in_progress","priority":1,"issue_type":"task","assignee":"claude-glm-glm47-alpha","created_at":"2026-06-10T03:53:44.914912586Z","updated_at":"2026-06-10T04:06:59.651877821Z","source_repo":".","compaction_level":0,"labels":["failure-count:1"]}