Index365 triage findings
Skill index365usa/agent-skills/skills/index365-triage-findings
Use when an index365 scan has findings and the user wants a plan before any code changes. Triggers: "what should I fix first", "prioritize the findings", "triage this audit", "turn the findings into a plan", "which of these matter". Produces a prioritized, file-mapped fix plan. It plans only, it does not edit code (use index365-apply-fix or index365-audit-and-fix to apply).From its SKILL.md
npx -y skills add index365usa/agent-skills --skill index365-triage-findingsAssembled 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
2.5 KB, as published. Nobody here has run it
index365 triage findings
Turn the latest run's findings into a prioritized, file-mapped plan. Planning only. Ideally you are inside the repository that serves the audited site, so findings can map to files.
Workflow
-
List findings worst-first (defaults to the latest run, no ids needed):
index365 findings --severity critical index365 findings --severity highFor a big report, save the full payload instead and rank over the file, never in context:
index365 report --savewrites it into the git-ignored.index365/dir; dispatch a subagent to rank findings from that file. -
Read the ones that matter.
index365 findings get <n> --jsonper candidate, using the ordinal from the table. Order by user impact and group bymarketingCategory.label. Do NOT group bycategory: it is a shared closed taxonomy with no Marketing Signal values, so it readsgeneralon every row. -
Map each finding to repo files. Use
affectedUrls+Glob/Grepto find the route/component/template that produces each affected URL. Note findings whoseaffectedUrlsdon't resolve to a file in this repo, those are out of scope for an in-repo fix. -
Output the plan (do not change code): an ordered list of finding ordinal +
findingId→ title → severity → target file(s) → the remediation from the finding. Cite the stablefindingIdin the plan, since ordinals are per-run positions whilefindingIdsurvives a re-scan for the after-fix comparison. Group into 1–3 reviewable batches, smallest safe diffs first. Flag anything that needs a human decision.
Shortcut via MCP prompts
If the index365 MCP server is connected, the triage_findings and prepare_pr_plan
prompts do steps 1–4 directly. Use them when you're in an MCP host; otherwise the CLI
flow above is equivalent.
Next
Apply the plan with index365-apply-fix (one finding at a time) or run the whole loop with index365-audit-and-fix.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.
Gives 0 of the 12 instructions most debug triage skills give
Counted across 1,020 of the 1,639 authors here whose files we hold, read 2026-09-06
- Find root cause before attempting any fixin 134 of 1020, across 118 files
- Create a failing test case before implementing a fixin 109 of 1020, across 95 files
- Read error messages and stack traces completelyin 102 of 1020, across 88 files
- Reproduce the issue consistently before investigatingin 90 of 1020, across 77 files
- Make the smallest possible change to test a hypothesisin 90 of 1020, across 76 files
- Trace data flow backward to find the sourcein 84 of 1020, across 70 files
- Form a single hypothesis before testingin 78 of 1020, across 64 files
- Implement only one fix at a timein 76 of 1020, across 63 files
- Question the architecture if three fixes failin 73 of 1020, across 59 files
- Add diagnostic instrumentation at component boundariesin 68 of 1020, across 56 files
- Compare broken code against working examplesin 68 of 1020, across 57 files
- Write a regression test before applying the fixin 62 of 1020, across 55 files
Said here and by no other author read
- List findings worst-first by severity
- Save full reports to the index365 directory
- Order findings by user impact
- Group findings by marketingCategory label
- Map findings to repository files using affectedUrls
- Identify findings out of scope for in-repo fixes
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once.