Bank Reconciliation

Perform bank reconciliation between bank statement and general ledger Excel files. Supports custom thresholds for amount matching and semantic description ma...

Security Scan

The skill's code and instructions match its stated purpose (reconciling two Excel files) and do not request unrelated credentials, network access, or elevated privileges.

Purpose & Capability

The name/description (bank reconciliation) align with the included script and SKILL.md. The code reads two XLSX files, performs key/description/amount matching, and writes a reconciliation XLSX — all coherent with the stated purpose. No unrelated binaries, env vars, or external services are required.

Instruction Scope

SKILL.md limits runtime actions to asking for input file paths/thresholds and running scripts/recon_logic.py. The script only reads the provided XLSX files and writes an output XLSX; it does not reference other system paths, environment variables, or network endpoints.

Install Mechanism

No install spec is present and the skill is instruction-only with one local Python script. Nothing is downloaded or written to system locations beyond the output file specified by the user.

Credentials

No credentials, secrets, or config paths are requested. The skill does not require environment variables or access to unrelated services.

Persistence & Privilege

always is false and the skill does not request persistent system presence or modify other skills or system-wide settings. It operates only when invoked and writes only the user-specified output file.

Assessment

This skill appears to do what it says: match transactions between two XLSX files and produce a results file. Before using on production data: (1) review the script yourself (it's local Python code) and run it on sample data; (2) note the script reads only sheet1.xml and certain cell types, so confirm it parses your Excel format correctly (shared strings or multiple sheets may not be handled); (3) back up originals and choose a distinct output filename to avoid overwriting; (4) the semantic matching uses a small hard-coded vendor list and simple heuristics — validate matches and thresholds manually for accuracy; (5) run in a restricted environment if you need to be extra cautious, although the script contains no network or credential-exfiltration behavior.

Bank Reconciliation Skill

This skill automates the process of matching bank statement records with general ledger entries.

Workflow

1. Input Verification: Identify the paths to bank_data.xlsx and gl_data.xlsx.
2. Threshold Configuration: Ask the user for a reconciliation threshold (e.g., "Allow matches within $1.00?"). Default is $0.00.
3. Execution: Run the scripts/recon_logic.py script to process the files.
4. Reporting: Provide the resulting recon_results.xlsx file which details:
   • Matched records from both sides.
   • Match basis (Amount, Description, or both).
   • Variances within the threshold.

Tool Usage

Run the reconciliation script using Python:

python3 scripts/recon_logic.py <bank_xlsx> <gl_xlsx> <output_xlsx> <threshold>

Description Matching Logic

The skill uses semantic matching to identify common vendor names, invoice IDs, or transaction references between the bank description and GL memo.