docs(token-audit): include figma fixes#58
Conversation
|
Warning Rate limit exceeded
Your organization is not enrolled in usage-based pricing. Contact your admin to enable usage-based pricing to continue reviews beyond the rate limit, or try again in 43 minutes and 50 seconds. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: ASSERTIVE Plan: Pro Run ID: 📒 Files selected for processing (1)
📝 WalkthroughWalkthroughRewrote the pf-token-auditor SKILL docs to replace step-based workflow with outcome- and rule-oriented guidance, require Figma deep links and explicit "Where" guidance, redefine finding/handoff formats (single-node links, multi-node "Affected nodes"), add PASS summary table, and specify interactive remediation contracts and figma-use execution constraints. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Suggested reviewers
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@plugins/design-tokens/skills/pf-token-auditor/SKILL.md`:
- Around line 133-160: Update the confirmation UI templates in SKILL.md so both
the initial numbered list and the per-finding "Update individually" prompt
include the missing evidence fields: add Node ID (e.g., include node-id: 123:456
next to each node name in the initial list), show the Figma source
value/variable and the CSS token target (e.g., `background-color` →
`background/color/primary/default`) inline, and append the token-description
evidence citation for each finding (a short token description string) in both
the “Update all” and “[1/{N}]” blocks; ensure the same fields appear for all
action options (“Update all”, “Update individually”, “Skip”, “Update”, “Skip”,
“Update remaining”) so approvals are fully informed.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: 46eb9af4-7de6-4cb9-a51f-e1adabf5e55a
📒 Files selected for processing (1)
plugins/design-tokens/skills/pf-token-auditor/SKILL.md
There was a problem hiding this comment.
Actionable comments posted: 3
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@plugins/design-tokens/skills/pf-token-auditor/SKILL.md`:
- Around line 1-4: The SKILL.md frontmatter for the pf-token-auditor skill lacks
the required safety flag; update the YAML frontmatter at the top of SKILL.md
(the block containing name: pf-token-auditor and description) to include
disable-model-invocation: true so the mutating skill that performs write
operations to Figma is marked as side-effect unsafe; ensure the new key is added
at the root of the frontmatter block alongside name and description and use the
exact literal disable-model-invocation: true.
- Around line 186-250: The README section titled "CRITICAL — Cursor:
`AskQuestion` is mandatory (not markdown bullets)" hard-codes Cursor's
AskQuestion schema; make it tool-agnostic by replacing the Cursor-specific JSON
and required field names with a neutral requirement: if the host exposes a
"clickable options" tool (any tool that accepts a prompt+options and returns a
selected option id), invoke that tool to present the three choices (update_all,
update_individually, skip) and fall back to a deterministic numbered-reply UI
("Reply 1, 2, or 3:") when no such tool exists; remove or generalize the exact
AskQuestion invocation examples and instead show a schema-agnostic example shape
(prompt, options[], option.id) and note the per-item variant similarly, ensuring
you still reference the option ids (update_all, update_individually, skip,
apply, provide_own, skip, apply_remaining) so callers can map them to their
tool.
- Around line 358-392: The "Execution via figma-use" section currently contains
host-specific implementation steps (e.g., figma.getNodeByIdAsync,
figma.variables.getLocalVariablesAsync,
figma.variables.setBoundVariableForPaint, direct mutation snippets and the
example return { mutatedNodeIds: [...] }) instead of high-level outcome
constraints; replace those procedural API examples with outcome-level
requirements that reference the same intent: require that each use_figma call
target specific node-ids from the audit, group fixes by operation type (variable
binding, composite replacement, value sync), return a list of mutated node IDs,
and include post-run validation (e.g., call get_metadata or get_screenshot) to
confirm mutations — remove concrete figma.* API calls and mutation code and
rewrite as goals/constraints (target node IDs, apply approved fix, verify
mutation, report mutatedNodeIds).
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: 07f4695b-e371-4ce6-b633-53e27741f942
📒 Files selected for processing (1)
plugins/design-tokens/skills/pf-token-auditor/SKILL.md
Based on Coderabbit feedback i made things more outcome focussed and less on the execution
There was a problem hiding this comment.
♻️ Duplicate comments (1)
plugins/design-tokens/skills/pf-token-auditor/SKILL.md (1)
1-4:⚠️ Potential issue | 🟠 MajorAdd side-effect safety flag to frontmatter.
This skill includes an edit path (“Optional: apply fixes in Figma”), so frontmatter should explicitly declare side effects with
disable-model-invocation: true.Proposed patch
--- name: pf-token-auditor description: Validate and bridge Figma design styles to PatternFly 6 design tokens. Use when auditing Figma designs against PatternFly tokens, validating token naming, translating Figma styles to composite tokens, or when the user mentions "token validation", "token audit", "design tokens", "Figma audit", "Figma variables", "token bridge", or "PF tokens". +disable-model-invocation: true ---As per coding guidelines: “Add
disable-model-invocation: truein SKILL.md frontmatter if the skill has side effects (creates issues, posts comments, deploys).”Also applies to: 125-171
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@plugins/design-tokens/skills/pf-token-auditor/SKILL.md` around lines 1 - 4, The SKILL.md frontmatter for the pf-token-auditor skill is missing the required side-effect safety flag; because this skill exposes an edit path ("Optional: apply fixes in Figma") you must add disable-model-invocation: true to the YAML frontmatter near the existing name: pf-token-auditor and description fields so the manifest explicitly declares side effects and prevents model-driven invocations.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Duplicate comments:
In `@plugins/design-tokens/skills/pf-token-auditor/SKILL.md`:
- Around line 1-4: The SKILL.md frontmatter for the pf-token-auditor skill is
missing the required side-effect safety flag; because this skill exposes an edit
path ("Optional: apply fixes in Figma") you must add disable-model-invocation:
true to the YAML frontmatter near the existing name: pf-token-auditor and
description fields so the manifest explicitly declares side effects and prevents
model-driven invocations.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: ASSERTIVE
Plan: Pro
Run ID: b833e809-0015-4ca1-88fb-bc5ff6a7c868
📒 Files selected for processing (1)
plugins/design-tokens/skills/pf-token-auditor/SKILL.md
2 participants
Added a step to leverage figma-use skill to implement changes that the audit reports.
Summary by CodeRabbit