Index of every page on this site: llms.txt
All skills
a2ui-project avatar

/a2ui-remediate-problem

@6bf4cc2
by A2UIa2ui-project/a2ui18k stars
1,430
Run it yourself

Remediates a specific recommendation from an A2UI compliance report issue, or resolves any general GitHub issue, by inspecting context, implementing minimal targeted fixes, verifying tests, creating a branch, and opening a developer-signed Pull Request. Use when asked to fix or remediate an A2UI compliance audit finding, recommendation, or repository issue.

  • 4 files
  • 22.6 KB
  • Updated 2 weeks ago
  • GitHub
Use this Skill: https://skilld.dev/gh/a2ui-project/a2ui/a2ui-remediate-problem

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈96 tokens for metadata: the name and description. ≈1.4k when used: this file. ≈704 more on demand in 1 file.

Description uses 4.8% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈384 characters, or 4.8%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

A2UI Issue & Compliance Remediation Skill

This skill guides an AI coding agent or engineer on how to remediate a specific recommendation from an A2UI compliance report or resolve any general repository issue and submit a clean, developer-signed Pull Request.


Workflow

Follow these steps to remediate an issue:

Step 1: Identify & Read Issue Context

  1. Determine the Target Issue:

    • If the user provided an issue number or URL in their request, extract ISSUE_NUMBER.
    • If no issue was specified, ask the user (prefer using the ask_question tool): "Which GitHub issue would you like to remediate? (Please provide the issue number or URL, e.g. 2391)"
  2. Retrieve Context & Determine Issue Type:

    • Run the helper script to inspect the issue:
      python3 .agents/skills/a2ui-remediate-problem/scripts/extract_recommendation.py "${ISSUE_NUMBER}" [RECOMMENDATION_INDEX] --json
    • Scenario A: Compliance Report with Multiple Recommendations:
      • If a recommendation index was provided (e.g. 2), the script extracts that item's details and priority.
      • If no recommendation index was specified, the script outputs all available recommendations in the issue. Ask the user (via ask_question) which recommendation item they would like to address.
      • Review the surrounding audit sections in the report (such as ## Codebase Blueprint Compliance Audit, ## Code & Documentation Sync Audit, or ## Test Quality & Assertions Audit) for detailed findings.
    • Scenario B: General Issue (Bug Report, Feature Request, or Non-Audit Issue):
      • If the issue is not a compliance report, the script returns the issue title and body as the problem statement to remediate.

Step 2: Formulate & Apply Remediation

  1. Identify the files and codebases referenced by the issue or recommendation.
  2. Carefully inspect the current source code, documentation, or blueprint files.
  3. Implement the minimal necessary change to remediate the reported problem.
    • CRITICAL RULE: Do not make unrelated changes, refactorings, or formatting edits outside the scope of the target issue or recommendation.
  4. Format modified files and ensure code style compliance:
    ./scripts/fix_format.sh
  5. If modifying tests or scripts, execute local tests across the affected packages to ensure no regressions were introduced.

Step 3: Choose Destination & Create Workspace / Branch

  1. Check with the user on where they would like the changes created and what branch name to use (prefer using the ask_question tool or equivalent interactive tool if available):
    • Recommended branch name: remediation/issue-${ISSUE_NUMBER}-${RECOMMENDATION_INDEX} (or remediation/issue-${ISSUE_NUMBER} for general issues).
    • Let the user choose whether to create a separate git worktree (using their preferred local worktree conventions), checkout a new branch in the current directory, or apply changes in place.
  2. Note the chosen branch name in BRANCH_NAME.
  3. Ensure the working tree is clean and up to date with upstream/main.

Step 4: Commit Changes

  1. Stage all modified files relevant to the remediation:
    git add -u
  2. Create a conventional commit referencing the issue number:
    • For compliance recommendations:
      git commit -m "fix(compliance): remediate issue #${ISSUE_NUMBER} recommendation ${RECOMMENDATION_INDEX}"
    • For general issues:
      git commit -m "fix(scope): remediate issue #${ISSUE_NUMBER} - ${CONCISE_SUMMARY}"

Step 5: Create Pull Request & Notify Issue (Optional / Upon Request)

  1. Check with the user whether they would like to submit a Pull Request (prefer using the ask_question tool or equivalent interactive tool if available):
    • Recommended default: Submit as a Draft PR (--draft) so CI checks run while allowing final review before requesting maintainer approvals.
    • Alternatively, submit directly as ready for review, or skip PR creation if the user wants to inspect changes locally first.
  2. If submitting a PR, generate a clear, descriptive PR title following Conventional Commits that concisely explains the specific fix being made (e.g. fix(swift): create missing top-level README in swift/core). Do NOT include generic issue or recommendation numbers in the PR title.
  3. Draft the PR description into a temporary file pr_description.md following pr-description-template.md.
  4. Push the branch to origin (or user's fork) and create the Pull Request against main:
    git push -u origin "${BRANCH_NAME}"
    PR_URL=$(gh pr create --draft \
      --repo a2ui-project/a2ui \
      --head "${BRANCH_NAME}" \
      --base main \
      --title "${PR_TITLE}" \
      --body-file pr_description.md)
    (Note: Omit --draft if the user prefers an immediate ready-for-review PR. If pushing from a fork, gh pr create automatically resolves the fork's head branch or accepts --head <username>:${BRANCH_NAME}.)
  5. Clean up the temporary file pr_description.md.
  6. Comment on the original issue to notify maintainers of the new PR:
    gh issue comment "${ISSUE_NUMBER}" \
      --repo a2ui-project/a2ui \
      --body "🤖 Automated remediation triggered! Created PR (${PR_URL}) on branch \`${BRANCH_NAME}\` to address recommendation ${RECOMMENDATION_INDEX}."

References

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No third-party reports yet.

Signed by skilld at 6bf4cc2. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 2 weeks ago
metadata
{
  "internal": true
}

README badge

README badge for a2ui-project/a2ui/a2ui-remediate-problem