mozempkandCopilot
383129f571
Recovered from crashed session (Node OOM). Repo contains full P0-P6 scaffold: plugin.json/marketplace.json, AGENTS.md, ADRs 0001-0006, lh CLI (init/index/graph/lane/run/memory/host/report/doctor), 10 .github/agents, 12 CLI skills, instructions, context7 mcp.json, and unit/e2e test suite. Fixed: run.mjs read --in-tokens/--out-tokens but tests and CLI docs use --input-tokens/--output-tokens, so telemetry totals were always 0. Now accepts both forms. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
1.3 KiB
1.3 KiB
applyTo, description
| applyTo | description |
|---|---|
| ** | Rules for reading, writing, compacting, and protecting lean harness memory shards. |
Memory Rules
Read memory
- Load
.agents/memory/INDEX.mdby default. - Do not load every shard at startup.
- Use
lh memory get --shard <name>only when needed. - Prefer compact shards over raw history.
Write memory
- Use
lh memory put --shard failuresfor recurring failures and fixes. - Use
lh memory put --shard correctionsfor user corrections and changed assumptions. - Use
lh memory put --shard insightsfor reusable design or codebase facts. - Use
lh memory put --shard conventionsfor durable project rules. - Use
lh memory put --shard quirksfor environment or tool behavior. - Keep entries short, dated, and source-linked.
- Record what happened, why it matters, and where it applies.
Protect secrets
- Never write secrets to memory.
- Never write tokens, keys, passwords, cookies, private keys, or credentials.
- Run
lh memory scanbefore saving risky content. - Redact sensitive values at source.
- Stop and ask for remediation if a secret is already present.
Compact memory
- Use
lh memory compactwhen shards grow noisy. - Preserve decisions, fixes, conventions, and source paths.
- Drop duplicate logs and stale speculation.