Skip to content

Releases: choiyounggi/groundwork

Release list

v0.2.0 — memory-loop: a memory lifecycle for your agent

v0.2.0 — memory-loop: a memory lifecycle for your agent Latest
Latest

Choose a tag to compare

View all tags
@choiyounggi choiyounggi released this 14 Jul 06:51
v0.2.0
591544f

groundwork grows a third axis: safety (guardrails) + quality (dev-loop) + continuity (memory-loop).

memory-loop (v0.1.0) — new plugin

A memory lifecycle layered on Claude Code's native file-based memory. Local-only, zero-config, MIT.

  • Save gate — confirm tier (long/short) and expiry with the user before anything is written; unverified guesses never enter long-term memory.
  • Tiered expiry — lapsed tier: short memories are moved to archived/ at session start, never deleted; the expiry day itself stays live (exclusive boundary).
  • Safe by default — files without a tier key (including everything that existed before install) are never touched; uninstalling leaves all memories intact.
  • Learning-review nudge — every N responses (configurable) the agent is reminded to distill habits, skills, and memories from recent work.
  • HABITS.md frame — turns corrections and incidents into standing behavior: positive practices (🟢) and hard lines (🛑), with merge-over-multiply and an escalation ladder to hooks/skills.
  • Identity — a one-time, declinable offer to set names for the user and the assistant (the assistant may pick its own name), injected as context each session. No CLAUDE.md edits, ever.
  • 23 bats tests + shellcheck, wired into CI.

guardrails and dev-loop are unchanged.

Install

/plugin marketplace add choiyounggi/groundwork
/plugin install memory-loop@groundwork

Then ask for "set up memory-loop" to run the guided setup.

MIT.

Assets 2
Loading

v0.1.0 — guardrails: safe-by-default Claude Code

v0.1.0 — guardrails: safe-by-default Claude Code

Choose a tag to compare

View all tags
@choiyounggi choiyounggi released this 13 Jul 05:22
v0.1.0
6f2bddc

First public release of groundwork — a safe-by-default Claude Code harness starter pack.

guardrails (v0.1.0)

A PreToolUse Bash guard + a redacted audit log. Local-only, zero-config, MIT.

  • Blocks supply-chain (piping a remote script into a shell), disk-destroying (dd/mkfs), and fork-bomb commands.
  • Asks before rm -rf, force-push, DROP/TRUNCATE, kubectl delete, credential/.env access, and cloud-resource deletion.
  • Redacted audit log — tokens/passwords masked before they're written.
  • Every rule is off / ask / block, per-repo or global, plus your own patterns.
  • Works in --dangerously-skip-permissions (yolo) mode — a PreToolUse deny still stops the command (verified headless).
  • /guardrails:self-test proves it live in ~10s, executing nothing.
  • bats tests + CI.

Bundled with dev-loop (a wiki-grounded plan→verify loop) in the same marketplace; the guard stands alone.

Install

/plugin marketplace add choiyounggi/groundwork
/plugin install guardrails@groundwork

MIT.

Loading