Skip to content
cccskills
FreePublished on GitHub

design-iteration

One-fix-at-a-time, approval-gated design fix recovery. Given a checklist of fixes (handover doc, CHECKLIST.md, or inline list), applies each individually, presents a diff, and waits for an explicit APPROVE / REJECT / NO CHANGES decision before committing. Trigger phrases: /design-iteration, "go through the checklist one by one", "fix these one at a time", "bucket recovery", "design fix pass".

The English summary is being prepared. Showing the original description.

See how to install

Files included (1)

  • SKILL.md4.9 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

/design-iteration

A one-fix-at-a-time, approval-gated design fix recovery skill. Given a checklist of fixes (a handover doc, CHECKLIST.md, or inline list), applies each fix individually, presents a diff, and waits for an explicit APPROVE / REJECT / NO CHANGES decision before committing or moving on. Never batches. Never self-merges.

When to invoke

  • /design-iteration — start or resume a fix pass on the current branch
  • "go through the checklist one by one"
  • "fix these one at a time, show me each before committing"
  • "bucket recovery" / "design fix pass" with a list of items

Invocation

/design-iteration [path/to/checklist-or-handover.md]

If no path is given, look for the active handover in:

  1. An explicit path in the conversation context
  2. founder/reports/design-review-<today>/HANDOVER*.md
  3. Ask the user to specify

Operating contract

The one rule that governs everything

One fix at a time. After each fix, STOP and present it for a decision.

  • APPROVE → commit that fix (one commit, one fix), push, move to next
  • REJECT → git restore touched files, move to next
  • NO CHANGES → fix was already present or unnecessary; touch nothing, note it, move to next

Never batch. Never "do them all and summarize." The checkpoint exists because these fixes touch files that may have been rewritten by a migration, and a wrong port can silently regress a shipped fix.

Loop for each fix

  1. Verify it is still missing. Re-run the check from the handover (grep, line read). If already present → NO CHANGES; report and skip.
  2. git show <ref-sha> (if provided) to see original intent. Do not blind-apply.
  3. Port the intent onto the current file(s) by hand. Do NOT paste stale patches wholesale when target files were rewritten since the source SHA.
  4. Read the edited file(s) back. Confirm:
    • No duplicated or merged lines
    • No new hardcoded hex (use design tokens; fallback to var(--token, fallback) only)
    • No banned patterns (gradient text, dead enum reads, client-side branch derivation if the project has an emitted-branch architecture)
  5. STOP. Present to the user:
    • Fix ID + description
    • Files changed
    • git diff output (or key hunks if very large)
    • One-line rationale: why this matches the intent without introducing regressions
    • Ask explicitly: APPROVE / REJECT / NO CHANGES?
  6. On decision:
    • APPROVE → git add <only those files> → commit with message fix(design): <fix-id> — <one-line summary> → git push origin <branch> → report SHA → next fix
    • REJECT → git restore <those files> → next fix
    • NO CHANGES → touch nothing → next fix
  7. Never proceed to the next fix before a decision on the current one.

After all fixes

Open one PR from the current branch to the target base branch. PR body must list:

  • Each fix + its outcome (approved / rejected / no-change)
  • Any Bucket-2 / "do not re-apply" items that were deliberately skipped, and why

Then stop — the human merges.

Branch + scope rules

  • Always work on the branch specified in the handover or by the user. Never create a new branch mid-pass without asking.
  • Respect CI branch-scope (e.g. design/ branches may only touch docs/ + *.md). When a fix crosses scope, surface the conflict and ask: split across branches, or add [skip-scope-check] to the commit message per the user's call.
  • Never push to main directly.

Presenting the diff

For each fix, show:

--- Fix N: <title> ---
Files: <list>
---
<git diff or key hunks>
---
Why this is safe: <one sentence>

APPROVE / REJECT / NO CHANGES?

If the diff is very large (e.g. 48-file sweep), show the full file list + one representative sample diff, then ask for approval before running the bulk operation.

Token discipline

  • Verify before editing (one read + grep). Edit. Verify after (read-back). That's the full cycle — do not re-read files already confirmed.
  • For bulk deletions (e.g. removing a dev tag from N files), use a single shell command rather than N individual edits, then confirm count = 0 with a follow-up grep.

Reference implementation

This skill was extracted from the 2026-07-20 Bucket-1 recovery session (design/ygg2-bucket1-recovery → dev/yggdrasil-ii-staging, PR #1241). Handover doc: founder/reports/design-review-2026-07-20/HANDOVER-bucket1-recovery.md. INSIGHTS source-of-truth: founder/reports/design-review-2026-07-20/INSIGHTS.md.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Measure CI churn cost for a pull request: how many commits and how many CI-compute seconds were burned on avoidable "fix the CI" push rounds. Use when you want to understand the efficiency cost of a PR's iteration cycle, get actionable pre-push check suggestions, or include a churn summary in a pipeline report. Triggers: "ci churn", "how much CI did we burn", "avoidable commits", "ci efficiency", "wasted CI time", "churn report", "/ci-churn". Called automatically by /fp-drift at pipeline close.

The English summary is being prepared. Showing the original description.

gaia-research/gaia-skill-tree23Updated Oct 17, 2026

Produce the rendered evidence a human-gated frontend PR owes its reviewer: capture each changed surface (plus the before state for anything modified, plus mobile width), then generate one self-contained local HTML page the founder can open and review without checking anything out. Use when a PR is gated by the "Frontend changes are HUMAN-GATED" section of CLAUDE.md, or when the user says "/design-gate", "screenshot evidence for this PR", "prep this for founder review", "build the design evidence page", "what does this look like before and after". The page comes from a template script, not hand-written markup. Output is session-local and never committed. Do NOT use for non-gated changes: typo/grammar/factual fixes in an unchanged block, dead link targets, alt/ARIA fixes with no visual change, cache-bust bumps, or refactors with no rendered diff.

The English summary is being prepared. Showing the original description.

gaia-research/gaia-skill-tree23Updated Oct 17, 2026

Playbook for correcting a named skill's star level to match its computed Trust Magnitude grade, with an independent post-calibration verification pass and a CLI-logged timeline event (never hand-written). Use when: recalibrating a skill's level, fixing a stale or drifted star rating, running a batch of `gaia dev calibrate` corrections, or confirming a past calibration actually landed. Triggers: "calibrate <skill>", "recalibrate the trust magnitude backlog", "promote/demote <skill> to N stars", "why is <skill> still N stars", "verify a calibration actually landed", "Hall of Heroes ranking is wrong".

The English summary is being prepared. Showing the original description.

gaia-research/gaia-skill-tree23Updated Oct 17, 2026

Close out a completed EPIC / sprint end-to-end: author a fact-verified meta-post announcing what shipped, archive the EPIC's handovers, snapshot orchestrator memory, gate the integration→main merge on CI-green + explicit human go, drive the version bump (merge-commit conventions that auto-sync classifies), and verify the release reached the registry/PyPI. Use when the user says "close out EPIC <n>", "wrap up the sprint", "ship the meta shift", "/epic-close", or asks to finalize an integration branch that is ready to merge to main. Generalizes the EPIC 1002 (Yggdrasil II) close-out. This skill orchestrates — it authors founder docs directly (superadmin) and delegates code/regen to workers; it NEVER merges to main without the human gate.

The English summary is being prepared. Showing the original description.

gaia-research/gaia-skill-tree23Updated Oct 17, 2026

Run this skill for Phase 3 of the evidence verification pipeline — the adversarial audit step. Use when you need to check the evidence data lake for bad data before ingestion: dead links, wrong URL formats (tree/ vs blob/), subjective wording ("elite", "high-quality"), stale migration notes, benchmark catalog misuse/vendor-claim leakage, or skills whose star evidence conflicts with classified evidence level. Triggers on phrases like: "audit the data lake", "adversarial check", "ev-adversarial-audit", "check for noise in evidence", "flag bad evidence", "run the audit phase", "quality check the by-type files", or any reference to Phase 3 of the pipeline.

The English summary is being prepared. Showing the original description.

gaia-research/gaia-skill-tree23Updated Oct 17, 2026

Phase 2B of the Gaia evidence verification pipeline. Verifies benchmark-result rows and benchmark-source candidates against registry/benchmark-sources.json before adversarial audit, link validation, catalog promotion, or ingestion. Use when running /ev-pipeline, checking benchmark-result evidence, reviewing benchmark-source candidates, or preparing benchmark rows for /gaia-ingest-batch.

The English summary is being prepared. Showing the original description.

gaia-research/gaia-skill-tree23Updated Oct 17, 2026

See all skills by gaia-research

Report a problem with this skill