CtrlK
BlogDocsLog inGet started
Tessl Logo
PluginPublic

adobe/aem-cloud-service

All AEM as a Cloud Service skills: component development, Workflow model design, development, triggering, launchers, debugging, and triaging, Dispatcher config authoring, advisory, incident response, performance tuning, and security hardening, Rapid Development Environment (RDE) deployment, log inspection, snapshots, and troubleshooting via `aio aem rde`, Jackrabbit Oak query/index tuning (property, Lucene index definitions), and extracting/inventorying JCR queries (XPath, SQL2, QueryBuilder) across a codebase.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Overview
Quality
Evals
Security
Files

troubleshooting.mdskills/code-assessment/references/

Troubleshooting

Decision tree for common failure paths. Execution steps live in runbook.md.

"Baseline compile failed"

Recorded in the report as compile.baseline=fail — the run continues to plan/report and can still apply after an explicit yes that acknowledges baseline failure. The skill does not fix unrelated pre-existing errors. If verification fails after apply, rollback still applies.

"Verification compile failed"

Edits reverted per edit_mode (git restore + branch delete, or in-place path list for manual revert). See .autofix/last-run.json.

Common causes: breaking API after a dependency upgrade, transitive conflicts, a wrong user-supplied target version. The skill does not auto-retry with lower versions.

"Git working tree not clean" (git mode only)

Does not block the run. The plan/summary warns that WIP and autofix edits may mix; rollback via git restore only affects paths the skill changed. Stash or commit first only if the user wants an isolated review.

"Not a git repository" / in-place run

Expected when edit_mode=in_place. No autofix/* branch; review modified paths in the IDE. Revert manually if verify failed.

"Edit skipped"

Reason prefixMeaningRecovery
literal-not-found: …Pom dependency triple not present in the fileRe-scan / re-inspect the pom
property-missing: … / property-value-mismatch: …Property shape changed or value differsRe-inspect the pom
property-not-referenced: …Property unused by the target dependencyLikely a false candidate
ambiguous-locator: …Non-unique pom matchFix manually
inject-ambiguous-field: …Whole file skipped per policySee ../inject-in-sling-model/recipe.md
discovery-no-match: …Discover mode found no antipattern hitsWiden scope or name the files explicitly

Re-running

Runs are independent. A fixed issue won't be re-found on the next scan, so re-running after a successful fix finds fewer (or no) candidates.

Known limitations

  • outdated-dependencies — literal <version> + same-pom ${property} only; target versions are user-supplied.
  • inject-in-sling-model — @Inject field injection in @Model only; constructor/setter @Inject and Felix SCR are not supported.
  • Compile only — no runtime Sling adaptation check; deploy to a dev instance before merge.
  • No remote git — no push/PR; in-place mode has no automatic file backup.

skills

README.md

tile.json