self-improve

self-improve — closing the loop on a convention library

A review comment that recurs is not a comment — it is an undocumented requirement. Once the same guidance appears often enough it belongs in the system, not in a human review. This skill turns that insight into a converging loop: mine the recurring signal → fold it into the generator/conventions → prove the change with a three-tier gate. As the library improves it produces fewer repeat comments → less signal → smaller changes → steady state (natural damping, not infinite mutation). When conventions change, the surge of new comments restarts the loop exactly where needed.

Adapted from Salesforce Engineering, "Closing the Loop: How to Build Self-Improving AI Systems with Automated Feedback Loops" (2026-07-17, forcedotcom/sf-skills).

When to Use This Skill (and When NOT to)

Use this skill for	Use a simpler approach for
A correction that has recurred ~3+ times across skills/PRs/sessions	A one-off fix — just fix it
Adding/revising a skill and wanting a real quality gate before release	A trivial typo/wording edit
Periodic "harden the library" / trigger-collision sweep	A single skill you already know is fine
Promoting a repeated preference into a durable rule	An ephemeral session fact

Start simple. Do not run the full loop for a single edit. The loop earns its cost only when the same signal repeats — that repetition is the whole trigger.

The loop

mine signal → apply (bounded) → three-tier gate → promote rule → measure → converge

1. Mine the signal. Collect recurring corrections (from feedback memories, PR review threads, repeated session corrections). Apply a frequency threshold: a pattern seen ~3+ times is a requirement, not a one-off. See references/rubric.md.
2. Apply, bounded. Fold the rule into the generator surface (skill body, a global CLAUDE.md convention, a template). Blast-radius caps: ≤5 improvements and ≤100 changed lines per cycle. Larger → split into cycles. Any regression in the gate aborts.
3. Three-tier gate (below). All three must pass before delivery.
4. Promote the rule to durable memory (a decision-memory store; see integration seams).
5. Deliver transparently. Consequential steps (auto-edit, push PR) pass a governance gate and produce a draft PR for human review, never an auto-merge.
6. Measure convergence. Track the decline in repeat-signal frequency across cycles (the direct efficacy metric). When signal dries up, stop; it restarts on the next convention change.