# BooCode — External Code Review & Lift Inventory

Last updated: 2026-05-20

This document tracks every open source repo BooCode references or lifts code from. Pin this so we don't lose attribution and don't re-evaluate the same projects twice.

BooCode is personal/single-user — license compatibility is non-blocking, but the License column is recorded so we don't accidentally inherit an obligation if BooCode ever goes public.

-----

## Reference repos

### Tier A — actively lifting from / running as sidecar

#### 1. sst/opencode (NEW Tier A as of 2026-05-20)

- **URL:** https://github.com/sst/opencode
- **License:** MIT
- **Language:** TypeScript (Effect-TS service-oriented)
- **What it is:** The coding agent Sam uses via Termius/Paseo. Also the source of every algorithm BooCode is porting through v1.15.
- **Why it matters:** opencode's `packages/opencode/src/session/` is the canonical reference implementation for every part of the inference layer BooCode is rebuilding. We lift the algorithms, not the Effect-TS plumbing.
- **Algorithms lifted so far:**
  - `session/compaction.ts` → v1.11.0 (shipped). `usable`, `isOverflow`, `select`, `buildPrompt` ported to plain TS. SUMMARY_TEMPLATE markdown skeleton verbatim.
  - `session/overflow.ts` → v1.11.0 (shipped). 20k `COMPACTION_BUFFER` constant.
- **Algorithms lifted (queued):**
  - `session/processor.ts` `DOOM_LOOP_THRESHOLD=3` → v1.11.6
  - `session/llm.ts` `experimental_repairToolCall` → v1.12 (hand-rolled), then v1.13 (via AI SDK)
  - `tool/truncate.ts` truncation + outputPath pattern → v1.12 (adapted: opaque id, not filesystem path)
  - `session/prompt.ts` `runLoop()` outer agent loop → v1.14
  - `permission/evaluate.ts` wildcard ruleset → v1.15
  - MCP client (transport, tools/list discovery, tools/call) → v1.15
- **What NOT to use:** Effect-TS service plumbing. Snapshot/patch system (for tool-edit revert; BooCoder territory if needed). The `experimental_native_runtime` (AI SDK fallback path). opencode's prompts.
- **Source tag:** `dev` branch on `sst/opencode`. Note: `anomalyco/opencode` is a rebranded mirror; use `sst/opencode` as canonical.

#### 2. nmakod/codecontext

- **URL:** https://github.com/nmakod/codecontext
- **License:** MIT
- **Language:** Go (single binary)
- **What it is:** AI-oriented codebase context map generator. Tree-sitter parsing across TS/JS/Go/C++/Swift/Python/Java/Rust/Dart/JSON/YAML. Generates `CLAUDE.md`-style structured overview. Bundled MCP server with 8 tools.
- **MCP tools exposed:** `get_codebase_overview`, `get_file_analysis`, `get_symbol_info`, `search_symbols`, `get_dependencies`, `watch_changes`, `get_semantic_neighborhoods` (git co-change patterns — no embeddings), `get_framework_analysis`.
- **Why it matters:** Solves the "architect needs a map" problem without embeddings.
- **How we use it:** Run as sidecar container in v1.12. Wire its MCP tools into BooCode's `inference/tools.ts` as static wrappers in v1.12, then re-wire via real MCP client when v1.15 ships.
- **What NOT to use:** Nothing. Clean fit.

#### 3. aimasteracc/tree-sitter-analyzer

- **URL:** https://github.com/aimasteracc/tree-sitter-analyzer
- **License:** MIT
- **Language:** Python, MCP server + CLI
- **What it is:** Local-first code context engine. Outline-first navigation, ripgrep-based impact trace, no embeddings. 17 languages. Claims 54-56% token reduction via TOON format.
- **MCP tools exposed:** `get_code_outline`, `trace_impact`, plus structural search/extract tools.
- **Why it matters:** Backup analyzer with a different response shape — outline-first scales better than codecontext's full dump on huge files. Impact trace is useful for "what calls this function" without a full graph build.
- **How we use it:** Lift the AST query patterns (`.scm` files) and the outline-first response shape. Can also run as a second MCP sidecar alongside codecontext.
- **What NOT to use:** Don't lift the TOON format if it conflicts with shadcn rendering — markdown stays.

#### 4. spirituslab/codesight

- **URL:** https://github.com/spirituslab/codesight
- **License:** check repo — assumed MIT-ish
- **Language:** TypeScript/Node
- **What it is:** Static code structure visualization. Symbol extraction, import resolution, call graphs. Detects circular dependencies and dead code (with documented false-positive caveats for `customElements.define()`, framework entry points, dynamic imports).
- **Why it matters:** Gives BooCode a `repo_health` tool — different from codecontext's "what is this" map. This is "what's wrong with this."
- **How we use it:** v1.16. Port the analyzer core (`analyze.mjs`). Call-graph builder + circular-dep + dead-code detectors into BooCode's `tools/repo_health.ts`. Drop the VS Code extension shell entirely.
- **What NOT to use:** The VS Code wrapper, the "idea layer" feature (requires Copilot or Claude Code wiring we don't want).

#### 5. Aider-AI/aider

- **URL:** https://github.com/Aider-AI/aider
- **License:** Apache-2.0
- **Language:** Python
- **What it is:** Git-native AI pair programmer CLI. Pioneered the tree-sitter repo-map + personalized PageRank approach.
- **Why it matters:** Authoritative source of per-language `tags.scm` query files. 60+ languages curated and battle-tested.
- **How we use it:** **Lift directly:** `aider/queries/tree-sitter-*.scm` — drop into BooCode's analyzer for any language codecontext or codesight don't cover natively.
- **What NOT to use:** Don't port `repomap.py` itself — codecontext supersedes it.

-----

### Tier B — patterns / partial lift

#### 6. continuedev/continue

- **URL:** https://github.com/continuedev/continue
- **License:** Apache-2.0
- **Language:** TypeScript
- **What it is:** IDE assistant framework. Full RAG pipeline, AST chunking, multi-provider LLM abstraction.
- **Why it matters:** One specific drop-in lift:
  1. `core/indexing/ignore.ts` — `DEFAULT_SECURITY_IGNORE_FILETYPES`. Three-tier matcher (basenames, extensions, prefixes). Going into BooCode's `pathGuard` to block analyzing `.env`, `.pem`, `id_rsa`, etc.
- **How we use it:** v1.11.7. Lift the ignore list, adapt to a `path.basename` + extension + prefix matcher.
- **What NOT to use:** `core/indexing/CodebaseIndexer.ts` and `LanceDbIndex.ts` — embedding-based, the path we walked away from.

#### 7. cline/cline

- **URL:** https://github.com/cline/cline
- **License:** Apache-2.0
- **Language:** TypeScript (VS Code extension)
- **What it is:** Autonomous coding agent. Pioneered plan/act mode and granular per-tool auto-approve.
- **Why it matters:** Pattern source for v1.15 (absorbed into the broader permissions work). Plan/act invariant: in plan mode, write tools hidden from the model's tool registry; in act mode, available but each individual tool can be approval-gated.
- **How we use it:** Lift the *pattern*, not the code. opencode's `permission/evaluate.ts` wildcard ruleset supersedes cline's mode-enum; cline contributes the conceptual framing (read-only invariant in BooCode v1.x).
- **What NOT to use:** Cline's VS Code-specific UI plumbing. The shape is wrong for our stack.

#### 8. plandex-ai/plandex

- **URL:** https://github.com/plandex-ai/plandex
- **License:** MIT
- **Language:** Go
- **What it is:** Terminal agent with a pending-changes sandbox. Edits never touch the filesystem until `/apply`. 2M token context.
- **Why it matters:** Reference architecture for BooCoder (v2.0). The "edits queue in a virtual layer, applied atomically" model is the right safety story for write tools.
- **How we use it:** Lift the data model: `pending_changes` table keyed by `(project_id, session_id, file_path)`, with diff content and apply/reject state. Lift the `diff` / `apply` / `rewind` UX vocabulary.
- **What NOT to use:** Plandex's 2M-context-window engineering. Our context is bounded by llama-swap.

#### 9. OpenHands/OpenHands

- **URL:** https://github.com/OpenHands/OpenHands
- **License:** MIT
- **Language:** Python
- **What it is:** Autonomous coding agent platform. V1 architecture is built on an append-only typed event log + Docker sandbox runtime.
- **Why it matters:** Two distinct patterns:
  1. Event-log architecture — superseded by v1.13's parts-table approach (which derives from opencode's part-message model). OpenHands event-log is conceptually similar but different shape.
  2. Sandbox runtime — per-session Docker container for write tools. Closes the `/opt:ro` mount risk.
- **How we use it:** v2.1. Lift the runtime container pattern (HTTP API inside container, BooCoder calls in). Don't port the Python implementation directly.
- **What NOT to use:** OpenHands' agent prompts, the full microagent system, the cloud deployment path. Event-log shape (use opencode-derived parts table instead).

-----

### Tier C — reference only / partial use / skip

#### 10. cortexkit/aft (actual repo path: ualtinok/aft)

- **URL:** https://github.com/ualtinok/aft
- **License:** check repo
- **Language:** Rust binary + TypeScript plugin
- **What it is:** Tree-sitter analysis tools delivered as a Rust binary, communicating with an OpenCode plugin via JSON-over-stdio. Warm-process pattern: one binary per project keeps parse trees in memory.
- **Why it matters:** The BridgePool transport model. If our `codecontext` tool calls get hot (agent loops calling it dozens of times per session), the warm-process pattern is faster than fork-per-call.
- **How we use it:** **Defer.** Profile first. Codecontext sidecar might be fast enough on its own. Revisit if tool-call latency becomes the bottleneck.
- **What NOT to use:** The opencode-plugin wrapper. Wrong integration surface.

#### 11. codeprysm/codeprysm

- **URL:** https://github.com/codeprysm/codeprysm
- **License:** check repo
- **Language:** Rust
- **What it is:** Graph-based code intelligence: tree-sitter parsing → node/edge graph in Qdrant, embeddings layered on top, MCP server exposes semantic search.
- **Why it matters:** Clean node/edge taxonomy: nodes = Container/Callable/Data; edges = CONTAINS/USES/DEFINES.
- **How we use it:** Lift the taxonomy *only* if we end up building our own graph instead of relying on codecontext. The embedding half is the trap we walked away from.
- **What NOT to use:** The Qdrant + embedding pipeline. Same anti-pattern as continue's indexer.

#### 12. DeepSourceCorp/globstar

- **URL:** https://github.com/DeepSourceCorp/globstar
- **License:** MIT
- **Language:** Go
- **What it is:** Static analysis toolkit for writing code checkers using tree-sitter S-expression queries. YAML interface for simple checkers, Go interface for complex multi-file checkers.
- **Why it matters:** Not for the architect tool. **Future use only.** If BooCoder ever grows a "verify before commit" lane, globstar checkers could be the verification engine: drop YAML checkers into `.globstar/`, run as a pre-apply gate.
- **How we use it:** Park. Not in any current version.
- **What NOT to use:** Don't try to use it as a codebase analyzer — it's a linter framework, wrong tool for the architect role.

#### 13. getpaseo/paseo

- **URL:** https://github.com/getpaseo/paseo
- **License:** AGPL-3.0
- **What it is:** WebSocket daemon ↔ client protocol for agent coordination. Already running in your stack (paseo dispatches Claude Code/opencode).
- **Why it matters:** Patterns for agent lifecycle, `--worktree` flag pattern, ECDH/NaCl security model.
- **How we use it:** Reference for BooCoder isolation (v2.0/v2.1). Note AGPL — fine for personal, blocks public distribution.
- **What NOT to use:** Don't vendor the source. Treat as a peer service.

#### 14. earendil-works/pi

- **URL:** https://github.com/earendil-works/pi
- **License:** MIT
- **What it is:** `@mariozechner/pi-agent-core` (tool loop + state machine) and `@mariozechner/pi-ai` (provider abstraction).
- **Why it matters:** If we ever want non-llama-swap inference (Anthropic, OpenAI, Mistral direct), pi-ai is the cleanest TypeScript provider abstraction available.
- **How we use it:** Defer. v2.x optional batch only.

#### 15. microsoft/agent-framework

- **URL:** https://github.com/microsoft/agent-framework
- **License:** MIT
- **What it is:** Workflow graphs for multi-agent coordination.
- **Why it matters:** Conceptual reference for far-future multi-agent orchestration.
- **How we use it:** Read the ADRs in `docs/decisions/`. Don't port code — implementation is Azure/Python/.NET-heavy.

#### 16. microsoft/autogen

- **URL:** https://github.com/microsoft/autogen
- **License:** MIT
- **What it is:** Earlier Microsoft multi-agent framework.
- **Why it matters:** Effectively sunsetting in favor of agent-framework.
- **How we use it:** Skip. Don't invest in evaluating further.

#### 17. open-webui/open-webui

- **URL:** https://github.com/open-webui/open-webui
- **License:** BSD-3
- **What it is:** Self-hosted LLM frontend.
- **Why it matters:** Python/Svelte, wrong stack. RAG pipeline only worth a read if BooLab needs improvement — unrelated to BooCode.
- **How we use it:** Skip for BooCode.

-----

## Lift catalog — what lands where

| Source repo | Specific artifact | License | BooCode destination | Version |
|---|---|---|---|---|
| `sst/opencode` | `session/compaction.ts` + `session/overflow.ts` algorithms | MIT | `services/compaction.ts` | **v1.11.0 ✅** |
| `sst/opencode` | `session/processor.ts` DOOM_LOOP_THRESHOLD pattern | MIT | `services/inference.ts` doom-loop guard | v1.11.6 |
| `continuedev/continue` | `core/indexing/ignore.ts` DEFAULT_SECURITY_IGNORE_FILETYPES | Apache-2.0 | Extend `path_guard.ts` exclusion list | v1.11.7 |
| `nmakod/codecontext` | Whole binary (sidecar) | MIT | New `codecontext` container, 8 MCP tools wired via static wrappers | v1.12 |
| `sst/opencode` | `session/llm.ts` experimental_repairToolCall pattern | MIT | `services/inference.ts` synthetic invalid-tool result | v1.12 |
| `sst/opencode` | `tool/truncate.ts` truncation + outputPath pattern (adapted: opaque id) | MIT | `services/truncate.ts` + `view_truncated_output` tool | v1.12 |
| `Aider-AI/aider` | `aider/queries/tree-sitter-*.scm` (60+ files) | Apache-2.0 | Fallback grammars for languages not covered by sidecars | v1.12 (fallback) |
| `sst/opencode` | `session/llm.ts` AI SDK adoption + alpha tool ordering | MIT | `services/inference.ts` rewrite | v1.13 |
| `sst/opencode` | Parts-message taxonomy (text, tool_call, tool_result, reasoning, step_start) | MIT | new `message_parts` table | v1.13 |
| `sst/opencode` | `session/prompt.ts` runLoop() outer agent loop | MIT | `services/inference.ts` step-based loop | v1.14 |
| `sst/opencode` | `agent.steps` per-agent step cap | MIT | AGENTS.md + agents.ts | v1.14 |
| `sst/opencode` | `permission/evaluate.ts` wildcard ruleset | MIT | new `permissions` table + matcher | v1.15 |
| `sst/opencode` | `mcp/index.ts` MCP client (SSE transport + tools/list + tools/call) | MIT | new `services/mcp/` module; codecontext re-wired through it | v1.15 |
| `cline/cline` | Plan/Act invariant (read-only mode pattern) | Apache-2.0 | absorbed into v1.15 permissions work | v1.15 |
| `spirituslab/codesight` | `analyze.mjs` — call graph, circular-dep, dead-code | MIT-ish | `apps/server/src/tools/repo_health.ts` | v1.16 |
| `plandex-ai/plandex` | `pending_changes` data model, diff/apply/rewind UX | MIT | New `pending_changes` table, BooCoder write-tool gating | v2.0 |
| `OpenHands/OpenHands` | Sandbox runtime pattern | MIT | New `boocoder` container, per-session Docker | v2.1 |
| `cortexkit/aft` (ualtinok/aft) | BridgePool warm-process JSON-stdio pattern | check | Optimization if profile shows fork overhead | Deferred |
| `codeprysm/codeprysm` | Node/edge taxonomy (Container/Callable/Data, CONTAINS/USES/DEFINES) | check | Reference only if we ever build our own graph | None |
| `DeepSourceCorp/globstar` | Whole toolkit | MIT | Future verify-before-commit gate for BooCoder | Parked |
| `earendil-works/pi` | `pi-ai` provider abstraction | MIT | Multi-provider LLM if pursued | v2.x optional |
| `microsoft/agent-framework` | Workflow graph concepts | MIT | Conceptual only | v3.x |

-----

## Decisions log

- **Embeddings dropped from BooCode** (May 2026). Replaced RAG with file-view tools + sidecar analyzers.
- **opencode promoted to Tier A** (2026-05-20). The compaction port (v1.11.0) made it clear opencode is not just "the agent Sam uses" — it's the canonical reference implementation for everything BooCode is rebuilding through v1.15. Five algorithms identified for lift (compaction, doom-loop, repairToolCall, runLoop, permission evaluate) plus truncate.ts and MCP client.
- **Source is `sst/opencode` `dev` branch.** `anomalyco/opencode` is a rebranded mirror; do not source from there.
- **Original Batch 11 (aider PageRank port) replaced** by codecontext sidecar approach.
- **Original Batch 12 (codebase indexer w/ Harrier) removed.** No embedding infrastructure.
- **Original Batch 13 (OpenHands event log) replaced** by v1.13 parts table (opencode pattern). Same outcome, different shape.
- **Original Batch 12 (cline plan/act mode) absorbed into v1.15** (opencode permission ruleset). Same outcome, wildcard rules instead of mode enum.
- **Aider's `repomap.py` port dropped.** Codecontext supersedes it. Aider contribution narrows to the `.scm` query files only.
- **Globstar role re-scoped.** Not an architect tool — parked for future verify-before-commit gate.
- **codeprysm role re-scoped.** Taxonomy reference only. Embedding half rejected.
- **AI SDK adoption deferred to v1.13.** Hand-roll opencode's repairToolCall pattern in v1.12 first.
- **`tool_choice='required'` confirmed supported** by llama-swap (qwen3.6-35b-a3b-mxfp4, 2026-05-20). Repair tool call is viable.
- **`anomalyco/sst` is a mirror, not a fork.** Same applies to `anomalyco/opencode`. Use canonical `sst/sst` and `sst/opencode` sources.