Rockachopa/Timmy-time-dashboard
1
0
Fork 2
Code Issues 6 Pull Requests Actions Packages Projects Releases Wiki Activity

[Autoresearch H1] Implement python -m timmy.cli learn Entry Point #907

New Issue
Closed
opened 2026-03-22 13:06:05 +00:00 by perplexity · 1 comment
perplexity commented 2026-03-22 13:06:05 +00:00
Collaborator
Copy Link

Parent

Part of #904 (Autoresearch Integration Proposal v2) — Action Item #6

Goal

Create a CLI entry point that triggers the autoresearch loop. The agent reads program.md (research direction file written by rockachopa) and iterates through experiments autonomously.

Implementation

  1. Add learn subcommand to src/timmy/cli.py
  2. On invocation:
    • Read program.md from repo root for research direction
    • Initialize SystemExperiment with target module and metric
    • Enter autoresearch loop: hypothesis → edit → tox → evaluate → commit/revert
    • Respect budget constraints (default 5 min per experiment)
  3. CLI flags:
    • --target: module or file to optimize
    • --metric: which metric function to use
    • --budget: time limit per experiment (minutes)
    • --max-experiments: cap on total experiments per run
    • --dry-run: show hypothesis without executing

Dependencies

  • SystemExperiment class must exist (sibling issue)
  • Aider integration already in toolkit (wire as execution engine)

Deliverable

python -m timmy.cli learn starts an autonomous improvement loop against the metric specified in program.md

Cross-references

  • #904 (Autoresearch Epic)
  • kimi-task backlog (~12 open refactoring tasks become initial experiment queue)

Owner

Kimi (account #5)

## Parent Part of #904 (Autoresearch Integration Proposal v2) — Action Item #6 ## Goal Create a CLI entry point that triggers the autoresearch loop. The agent reads `program.md` (research direction file written by rockachopa) and iterates through experiments autonomously. ## Implementation 1. Add `learn` subcommand to `src/timmy/cli.py` 2. On invocation: - Read `program.md` from repo root for research direction - Initialize `SystemExperiment` with target module and metric - Enter autoresearch loop: hypothesis → edit → tox → evaluate → commit/revert - Respect budget constraints (default 5 min per experiment) 3. CLI flags: - `--target`: module or file to optimize - `--metric`: which metric function to use - `--budget`: time limit per experiment (minutes) - `--max-experiments`: cap on total experiments per run - `--dry-run`: show hypothesis without executing ## Dependencies - SystemExperiment class must exist (sibling issue) - Aider integration already in toolkit (wire as execution engine) ## Deliverable `python -m timmy.cli learn` starts an autonomous improvement loop against the metric specified in program.md ## Cross-references - #904 (Autoresearch Epic) - kimi-task backlog (~12 open refactoring tasks become initial experiment queue) ## Owner Kimi (account #5)
claude was assigned by Rockachopa 2026-03-22 23:32:22 +00:00
claude added the harnessheartbeatp0-critical labels 2026-03-23 13:52:46 +00:00
claude commented 2026-03-23 23:07:07 +00:00
Collaborator
Copy Link

PR created: http://143.198.27.163:3000/rockachopa/Timmy-time-dashboard/pulls/1240

Implemented the timmy learn autoresearch entry point:

  • SystemExperiment class (autoresearch.py) — encapsulates hypothesis → Aider edit → tox → evaluate → commit/revert with configurable target, metric, and budget. Supports unit_pass_rate and coverage (higher-is-better) alongside val_bpb and custom metrics.
  • timmy learn command (cli.py) — flags: --target, --metric, --budget, --max-experiments, --dry-run, --tox-env, --model. Reads program.md for research direction; degrades gracefully when file is absent or Aider is not installed.
  • program.md — starter research direction template in repo root.
  • 21 new unit tests — all 400 unit tests pass.
PR created: http://143.198.27.163:3000/rockachopa/Timmy-time-dashboard/pulls/1240 Implemented the `timmy learn` autoresearch entry point: - **`SystemExperiment` class** (`autoresearch.py`) — encapsulates hypothesis → Aider edit → tox → evaluate → commit/revert with configurable target, metric, and budget. Supports `unit_pass_rate` and `coverage` (higher-is-better) alongside `val_bpb` and custom metrics. - **`timmy learn` command** (`cli.py`) — flags: `--target`, `--metric`, `--budget`, `--max-experiments`, `--dry-run`, `--tox-env`, `--model`. Reads `program.md` for research direction; degrades gracefully when file is absent or Aider is not installed. - **`program.md`** — starter research direction template in repo root. - **21 new unit tests** — all 400 unit tests pass.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
222-epic
Workshop: Timmy as Presence (Epic #222)
 actionable
Has a concrete code/config task extracted
 assigned-claude
Issue currently assigned to Claude agent — do not assign to another agent
 assigned-gemini
Issue currently assigned to Gemini agent — do not assign to another agent
 assigned-groq
assigned-kimi
Issue currently assigned to Kimi agent — do not assign to another agent
 assigned-manus
Issue currently assigned to Manus agent — do not assign to another agent
 claude-ready
consolidation
Part of a consolidation epic
 deprioritized
Keep open but not blocking P0 work
 deprioritized
Keep open but not blocking P0 work
 duplicate
Duplicate of another issue
 gemini-review
Auto-generated by Gemini, needs relevance review
 groq-ready
harness
Core product: agent framework, heartbeat, inference, memory
 heartbeat
Harness: Agent heartbeat loop
 inference
Harness: Inference and model routing
 infrastructure
Supporting stage: dashboard, CI/CD, deployment, DNS
 kimi-ready
Scoped and ready for Kimi to pick up
 memory-session
Harness: Memory and session crystallization
 morrowind
Harness: Morrowind embodiment
 needs-design
Needs architectural design before implementation
 needs-extraction
Philosophy with unextracted engineering work
 p0-critical
Priority 0: Must fix now
 p1-important
Priority 1: Important, next sprint
 p2-backlog
Priority 2: Backlog, do when time permits
 philosophy
Philosophical foundation — informs architecture decisions
 rejected-direction
Closed: rejected or superseded direction
 seed:know-purpose
Three Seeds: KNOW YOUR PURPOSE
 seed:serve-real
Three Seeds: SERVE THE REAL
 seed:tell-truth
Three Seeds: TELL THE TRUTH
 sovereignty
Harness: Sovereignty stack
No Label harness heartbeat p0-critical
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
Timmy claude grok groq kimi manus perplexity Rockachopa
No Assignees claude
2 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Rockachopa/Timmy-time-dashboard#907
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?