agentsclimarketplace

Index365 audit and fix

Skill index365usa/agent-skills/skills/index365-audit-and-fix

Use when the user wants their coding agent to audit their site with index365 AND fix what it finds in the current repo, not just report it. Triggers: "make my site AI-ready", "audit and fix my site", "fix everything index365 finds", "improve my index365 score", or the user pastes a report and says "act on this". This is the full autonomous loop. For a single finding use index365-apply-fix; to only read, use index365-read-report.From its SKILL.md

Install
npx -y skills add index365usa/agent-skills --skill index365-audit-and-fix

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

3.9 KB, as published. Nobody here has run it

index365 audit and fix

Run an index365 scan, turn findings into a prioritized plan, apply the fixes that live in THIS repo, and re-scan to confirm the score actually moved. The agent fixes; index365 supplies the findings and per-finding remediation. This is not a summary, it fixes. You need to be in the repository that serves the audited site (so affectedUrls map to files); no ids are carried between steps, every read defaults to the latest run.

Workflow (copy this checklist into your reply; check off as you go)

  • 1. Scan and wait: index365 scan https://yoursite.com --yes --idempotency-key "baseline-<date>". Marketing Signal is the only startable product, so there is nothing to choose. --yes creates the project on a new domain without a prompt. Paid action (one Page Check, 10 credits); the idempotency key makes a retry safe. Record the baseline score and its band from the score card.
  • 2. Read + prioritize: index365 findings --severity critical then high for the worst items. For a big report, index365 report --save writes the full payload into the git-ignored .index365/ dir; delegate ranking to a subagent over the FILE, never paste a full report into context. Prefer each finding's machine-readable agentActions over re-deriving the fix. (This is index365-triage-findings.)
  • 3. Apply in-repo fixes: per top finding, index365 findings get <n> --json (ordinal from the table); map affectedUrls → file with Glob/Grep. If the user requested preview or approval before application, show a proposed patch and STOP without editing. Otherwise, the direct audit-and-fix request authorizes one logical Edit per finding; show the actual diff. (This is index365-apply-fix, repeated.) STOP and ask if a fix exceeds the finding, or affectedUrls doesn't resolve to a file; skip out-of-repo findings. Note each fixed finding's stable findingId for the comparison in step 4.
  • 4. Re-scan and verify: repeat the step-1 command with the same URL and --product but a NEW stable key, e.g. --idempotency-key "verify-batch-1"; reuse that new key only when retrying this verification. Never reuse the baseline key, which would replay the baseline run. Compare the new score and index365 findings against the baseline, matching by stable findingId (ordinals shift between runs). A fix that doesn't move the score or clear its finding is not done, re-investigate.
  • 5. Report the delta: baseline → new score (index365 results yoursite.com shows both runs side by side), the findingIds fixed, applied-but-unmoved (re-investigate these), and out-of-repo findings with each humanUrl so the user can handle them in the dashboard.

Output

Write saved reports and artifacts to .index365/ (git-ignored). Field reference: references/finding-schema.md.

Common mistakes

  • Summarizing instead of fixing (that's index365-read-report).
  • Pasting the whole saved report into context instead of reading the file with jq/grep.
  • Re-deriving a fix when agentActions already specifies it.
  • Calling step 4 done without comparing scores against the baseline.
  • Editing files for findings whose affectedUrls are DNS/hosting/infra (not in this repo).

What ships with it: 1 file

4.1 KB alongside SKILL.md

references/

Gives 0 of the 12 instructions most audit compliance skills give

Counted across 960 of the 1,589 authors here whose files we hold, read 2026-09-06

  • Read product marketing context before asking questionsin 29 of 960, across 11 files
  • Rank findings by severityin 29 of 960, across 22 files
  • Generate audit reportin 22 of 960
  • Run the audit scriptin 20 of 960, across 19 files
  • Generate a prioritized action plan reportin 19 of 960, across 11 files
  • Ensure one H1 per pagein 15 of 960, across 5 files
  • Ensure sitemap exists and is accessiblein 14 of 960, across 4 files
  • Verify alt text on all imagesin 12 of 960, across 3 files
  • Determine the audit scope before startingin 12 of 960, across 4 files
  • Verify important pages allowed in robots.txtin 11 of 960, across 2 files
  • Detect business type from homepage signalsin 11 of 960, across 7 files
  • Delegate specialized tasks to subagentsin 11 of 960, across 7 files

Said here and by no other author read

  • Run an index365 scan
  • Apply in-repo fixes
  • Report the delta

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once.

Keep looking

Skills are one crate of 613,297. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.