ypi

ypi — a recursive coding agent built on Pi.

Named after the Y combinator from lambda calculus — the fixed-point combinator that enables recursion. ypi is Pi that can call itself. (rpi already has another connotation.)

Inspired by Recursive Language Models (RLM), which showed that an LLM with a code REPL and a llm_query() function can recursively decompose problems, analyze massive contexts, and write code — all through self-delegation.

The Idea

Pi already has a bash REPL. We add one function — rlm_query — and a system prompt that teaches Pi to use it recursively. Each child gets its own jj workspace for file isolation. That's the whole trick.

┌──────────────────────────────────────────┐
│  ypi (depth 0)                           │
│  Tools: bash, rlm_query                  │
│  Workspace: default                      │
│                                          │
│  > grep -n "bug" src/*.py                │
│  > sed -n '50,80p' src/app.py \          │
│      | rlm_query "Fix this bug"          │
│            │                             │
│            ▼                             │
│    ┌────────────────────────────┐        │
│    │  ypi (depth 1)            │        │
│    │  Workspace: jj isolated   │        │
│    │  Edits files safely       │        │
│    │  Returns: patch on stdout │        │
│    └────────────────────────────┘        │
│                                          │
│  > jj squash --from <child-change>       │
│  # absorb the fix into our working copy  │
└──────────────────────────────────────────┘

Using ypi

Install

# npm (global)
npm install -g ypi

# or run without installing
npx ypi "What does this repo do?"
bunx ypi "What does this repo do?"

# or curl
curl -fsSL https://raw.githubusercontent.com/rawwerks/ypi/master/install.sh | bash

# or manual
git clone https://github.com/rawwerks/ypi.git && cd ypi
git submodule update --init --depth 1
export PATH="$PWD:$PATH"

Run

# Interactive
ypi

# One-shot
ypi "Refactor the error handling in this repo"

# Different model
ypi --provider anthropic --model claude-sonnet-4-5-20250929 "What does this codebase do?"

How It Works

Three pieces (same architecture as Python RLM): | Piece | Python RLM | ypi | |---|---|---| | System prompt | RLM_SYSTEM_PROMPT | SYSTEM_PROMPT.md | | Context / REPL | Python context variable | $CONTEXT file + bash | | Sub-call function | llm_query("prompt") | rlm_query "prompt" | Recursion: rlm_query spawns a child Pi process with the same system prompt and tools. The child can call rlm_query too:

Depth 0 (root)    → full Pi with bash + rlm_query
  Depth 1 (child) → full Pi with bash + rlm_query, own jj workspace
    Depth 2 (leaf) → full Pi with bash, but no rlm_query (max depth)

File isolation with jj: Each recursive child gets its own jj workspace. The parent's working copy is untouched. Review child work with jj diff -r <change-id>, absorb with jj squash --from <change-id>.

Why It Works

The design has three properties that compound:

1. Self-similarity — Every depth runs the same prompt, same tools, same agent. No specialized "scout" or "planner" roles. The intelligence is in decomposition, not specialization. The system prompt teaches one pattern — size-first → search → chunk → delegate → combine — and it works at every scale.

2. Self-hosting — The system prompt (SECTION 6) contains the full source of rlm_query. The agent reads its own recursion machinery. When it modifies rlm_query, it's modifying itself. This isn't a metaphor — it's the actual execution model.

3. Bounded recursion — Five concentric guardrails (depth limit, PATH scrubbing, call count, budget, timeout) guarantee termination. The system prompt also installs cognitive pressure: deeper agents are told to be more conservative, preferring direct action over spawning more children.

4. Symbolic access — Anything the agent needs to manipulate precisely is a file, not just tokens in context. $CONTEXT holds the data, $RLM_PROMPT_FILE holds the original prompt, and hashline provides line-addressed edits. Agents grep/sed/cat instead of copying tokens from memory.

Guardrails

| Feature | Env var | What it does | |---------|---------|-------------| | Budget | RLM_BUDGET=0.50 | Max dollar spend for entire recursive tree | | Timeout | RLM_TIMEOUT=60 | Wall-clock limit for entire recursive tree | | Call limit | RLM_MAX_CALLS=20 | Max total rlm_query invocations | | Model routing | RLM_CHILD_MODEL=haiku | Use cheaper model for sub-calls | | Depth limit | RLM_MAX_DEPTH=3 | How deep recursion can go | | jj disable | RLM_JJ=0 | Skip workspace isolation | | Plain text | RLM_JSON=0 | Disable JSON mode (no cost tracking) | | Tracing | PI_TRACE_FILE=/tmp/trace.log | Log all calls with timing + cost |

The agent can check spend at any time:

rlm_cost          # "$0.042381"
rlm_cost --json   # {"cost": 0.042381, "tokens": 12450, "calls": 3}

Pi Compatibility

ypi is a thin layer on top of Pi. We strive not to break or duplicate what Pi already does:

| Pi feature | ypi behavior | Tests | |---|---|---| | Session history | Uses Pi's native ~/.pi/agent/sessions/ dir. Child sessions go in the same dir with trace-encoded filenames. No separate session store. | G24–G30 | | Extensions | Passed through to Pi. Children inherit extensions by default. RLM_EXTENSIONS=0 disables. | G34–G38 | | System prompt | Built from SYSTEM_PROMPT.md + rlm_query source, written to a temp file, passed via --system-prompt (file path, never inlined as shell arg). | T8–T9 | | -p mode | All child Pi calls run non-interactive (-p). ypi never fakes a terminal. | T3–T4 | | --session flag | Used when RLM_SESSION_DIR is set; --no-session otherwise. Never both. | G24, G28 | | Provider/model | Never hardcoded. ypi and rlm_query use Pi's defaults unless the user sets RLM_PROVIDER/RLM_MODEL. | T14, T14c |

If Pi changes how sessions or extensions work, our guardrail tests should catch it.

Contributing

Project Structure

ypi/
├── ypi                    # Launcher: sets up env, starts Pi as recursive agent
├── rlm_query              # The recursive sub-call function (Pi's analog of rlm llm_query())
├── SYSTEM_PROMPT.md       # Teaches the LLM to be recursive + edit code
├── AGENTS.md              # Meta-instructions for the agent (read by ypi itself)
├── Makefile               # test targets
├── tests/
│   ├── test_unit.sh       # Mock pi, test bash logic (no LLM, fast)
│   ├── test_guardrails.sh # Test guardrails (no LLM, fast)
│   └── test_e2e.sh        # Real LLM calls (slow, costs ~$0.05)
├── pi-mono/               # Git submodule: upstream Pi coding agent
└── README.md

Version Control

This repo uses jj for version control. Git is only for GitHub sync.

jj status                    # What's changed
jj describe -m "message"     # Describe current change
jj new                       # Start a new change
jj bookmark set master       # Point master at current change
jj git push                  # Push to GitHub

Never use git add/commit/push directly. jj manages git under the hood.

Testing

make test-fast    # 54 tests, no LLM calls, seconds
make test-e2e     # Real LLM calls, costs ~$0.05
make test         # Both

Before any change to rlm_query: run make test-fast. After: run it again. rlm_query is a live dependency of the agent's own execution — breaking it breaks the agent.

History

ypi went through four approaches before landing on the current design:

1. Tool-use REPL (exp 010/012) — Pi's completeWithTools(), ReAct loop. 77.6% on LongMemEval.
2. Python bridge — HTTP server between Pi and Python RLM. Too complex.
3. Pi extension — Custom provider with search tools. Not true recursion.
4. Bash RLM (rlm_query + SYSTEM_PROMPT.md) — True recursion via bash. Current approach.

The key insight: Pi's bash tool is the REPL. rlm_query is llm_query(). No bridge needed.

See Also

• Pi coding agent — the underlying agent
• Recursive Language Models — the library that inspired this
• rlm-cli — Python RLM CLI (budget, timeout, model routing)