Commit graph

4 commits

Author SHA1 Message Date
johnfrum1234
fabae74c19 Add context builder, agent runner, and pipeline executor
core/context_builder.py:
  build_context() — assembles role-specific context from DB.
  PM gets everything; debugger gets gotchas/workarounds; reviewer
  gets conventions only; tester gets minimal context; security
  gets security-category decisions.
  format_prompt() — injects context into role templates.

agents/runner.py:
  run_agent() — launches claude CLI as subprocess with role prompt.
  run_pipeline() — executes multi-step pipelines sequentially,
  chains output between steps, logs to agent_logs, creates/updates
  pipeline records, handles failures gracefully.

agents/specialists.yaml — 8 roles with tools, permissions, context rules.
agents/prompts/pm.md — PM prompt for task decomposition.
agents/prompts/security.md — security audit prompt (OWASP, auth, secrets).

CLI: kin run <task_id> [--dry-run]
  PM decomposes → shows pipeline → executes with confirmation.

31 new tests (15 context_builder, 11 runner, 5 JSON parsing).
92 total, all passing.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-15 14:03:32 +02:00
johnfrum1234
e5444114bd Fix bootstrap: deep scan, CLAUDE.md fallback, noise filtering
1. Tech stack: recursive file search (depth 3) + CLAUDE.md text fallback
   when config files are on remote server (detects nodejs, postgresql, etc.)
2. Modules: scan */src/ patterns in top-level dirs (frontend/src/, backend-pg/src/)
3. Decisions: filter out unrelated sections (Jitsi, Nextcloud, Prosody, GOIP),
   filter noise (commit hashes, shell commands, external service paths).
   Noise filtering also applied to Obsidian decisions.

Tested on vdolipoperek: 4 tech, 5 modules, 9 clean decisions, 24 Obsidian tasks.
61 tests, all passing.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-15 13:37:42 +02:00
johnfrum1234
da4a8aae72 Add bootstrap command — auto-detect project stack, modules, decisions
kin bootstrap <path> --id <id> --name <name> [--vault <path>]

Detects: package.json, requirements.txt, go.mod, config files → tech_stack.
Scans src/app/lib/frontend/backend dirs → modules with type detection.
Parses CLAUDE.md for GOTCHA/WORKAROUND/FIXME/ВАЖНО → decisions.
Scans Obsidian vault for kanban tasks, checkboxes, and decisions.
Preview before save, -y to skip confirmation.
18 bootstrap tests, 57 total passing.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-15 13:29:01 +02:00
johnfrum1234
432cfd55d4 Add CLI (cli/main.py) — click-based interface for all core operations
Commands: project (add/list/show), task (add/list/show),
decision (add/list), module (add/list), status, cost.
Auto-generated task IDs (PROJ-001). DB at ~/.kin/kin.db or $KIN_DB.
pyproject.toml with `kin` entry point. 18 CLI tests, 39 total passing.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-15 13:20:57 +02:00