Skip to content

Ronak1167/aura-vega-tv / .agents/skills/accounting-software-selection

accounting-software-selection

Verified

Scores shortlisted accounting packages against 57 evidence-backed fields, emitted as CSV, SQL, JSON Schema or Notion on request. Use for choosing accounting software.

FileKindSize
SKILL.mdskill md29.1 KB
references/build-artifacts.mdmarkdown18.8 KB
references/common-pitfalls.mdmarkdown3.5 KB
references/field-reference.mdmarkdown8.9 KB
references/select-options.mdmarkdown5.0 KB

references/common-pitfalls.md

View on GitHub
# Accounting Software Selection: Common Pitfalls

## Common Pitfalls

- **Problem:** a static mapping is described as a completed workspace build.
  **Solution:** deliver manual mappings without a connection; claim a live change only
  after the authorized tool operation succeeds.
- **Problem:** treated the average score as the answer.
  **Solution:** an average hides disagreement. Keep each evaluator's assessment separate and show the spread.
- **Problem:** let a strong score in one category hide a failed Must-have.
  **Solution:** flag `Deal-breaker: Yes` on the failed requirement, whatever the total says.
- **Problem:** wrote `1 Missing` for a capability nobody had got round to testing.
  **Solution:** `Untested`. Nobody looking is a gap in the evaluation, not a finding about the product.
- **Problem:** filled a capability field from a brochure.
  **Solution:** record it only from documentation, a demonstration, a trial, a test transaction or the quotation, and name the source and its date in `Evidence/Source`.
- **Problem:** wrote `Deal-breaker: No` on a candidate nobody had finished testing.
  **Solution:** `Unknown` until every Must-have has been tested. `No` asserts something nobody has checked.
- **Problem:** wrote a score with no evidence, or evidence with no date.
  **Solution:** `Evidence/Source` carries what kind of source it was and when it was gathered; a score nobody can trace is deleted, not kept.
- **Problem:** averaged two evaluators into one rating.
  **Solution:** keep each view on its own record or in their own words. The spread is the finding.
- **Problem:** scored security and backup/recovery as one number when the demonstration disagreed.
  **Solution:** `Data Backup & Security` is one field, so write the disagreement in `Notes` instead of averaging it away.
- **Problem:** used one `Recommendation` column for a decision, a reason and a caveat.
  **Solution:** `Evaluation Status`, `Selection Decision`, `Rejection Reason` and `Deal-breaker` are four fields for exactly that reason. Do not merge them.
- **Problem:** made the three-year total work by adjusting a cost line.
  **Solution:** record the difference in `Notes` and name the unverified line. A total that needs a component changed to balance is not a total.
- **Problem:** wrote a plausible price into an empty cost cell so the row looked finished.
  **Solution:** leave the cell empty and put `Unknown` in `Notes`. A total with a guessed line in it is a fabricated price.
- **Problem:** assumed NPR because the business is in Nepal.
  **Solution:** NPR is the default only. If the user names another currency, use theirs and record it beside the quotation.
- **Problem:** took `maybe` or `same` as an answer to a multiple-choice question.
  **Solution:** re-ask as an explicit choice, and keep only the part of a partial answer that was actually given.
- **Problem:** opened with the shortlist question when the request had already named the packages.
  **Solution:** read the request first and ask only what would change the evaluation.
- **Problem:** built a go-live checklist when only the comparison was asked for.
  **Solution:** build what was requested; mention the checklist and the parent skill separately.
- **Problem:** asked all the questions in one message.
  **Solution:** ask one, wait, and drop any the first answer already covered.
- **Problem:** all four artifacts drift apart.
  **Solution:** derive all four from the Field Reference, never by hand.
- **Problem:** Notion import shows every column as Text.
  **Solution:** that is expected. Apply the property mapping table once, after import.

deprecation-and-migrationaddyosmani/agent-skillsManages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to another. Use when migrating a database schema in production, such as renaming or dropping a column without downtime (expand/contract). Use when deciding whether to maintain or sunset existing code.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
performance-optimizationaddyosmani/agent-skillsOptimizes application performance across frontend, backend, queries, and databases. Use when performance requirements exist, when you suspect performance regressions, when Core Web Vitals or load times need improvement, when N+1 query patterns need fixing, or when profiling reveals bottlenecks.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
agent-cost-reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
agent-cost-reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json.Stars: 118kUpdated Claude CodeCodexVerifiedInstall