Tachograph Triage & Root-Cause Prompts (UK)

Triages tachograph infringements, identifies common patterns, and outputs “what to check next” prompts and weekly review notes. USE WHEN doing weekly tacho/WTD reviews.

Security Scan

Instruction-only triage helper for UK tachograph/WTD reviews; its required inputs and included guidance match the described purpose and it requests no credentials, installs, or external endpoints.

Purpose & Capability

The name and description (triaging tachograph infringements and producing weekly review notes) align with the SKILL.md and the bundled reference/playbook files. There are no unrelated credentials, binaries, or config paths requested.

Instruction Scope

Runtime instructions are narrowly scoped to normalising infringement records, pattern-scanning against the included checklist, producing per-driver prompts and a weekly pack, and stopping to ask when data is missing. The instructions do not direct the agent to read unrelated system files, access external endpoints, or exfiltrate data.

Install Mechanism

No install spec or code files — instruction-only. Nothing is downloaded or written to disk by an installer in the skill bundle.

Credentials

The skill declares no required environment variables, credentials, or config paths. The workflow expects user-supplied infringement data (CSV/PDF extracts) which is appropriate for the stated purpose.

Persistence & Privilege

always is false and the skill is user-invocable; it does not request permanent presence or attempt to modify other skills or system settings. Autonomous invocation is allowed by platform default but the skill itself does not request elevated persistence.

Assessment

This skill is an instruction-only template for triaging tachograph infringements and appears internally consistent. Before using it: (1) Validate outputs on sample, non-sensitive data to confirm the triage logic matches your policies; (2) Do not feed full PII-laden PDFs or personnel records into any third-party agent unless you trust the execution environment and have appropriate data-protection approvals; (3) Ensure any RAG/escalation thresholds referenced are configured to match your organisation’s policy (the skill flags where history is missing but relies on you to supply counts/outcomes); (4) If a future version adds code, external endpoints, or requests environment credentials, stop and reassess — that would materially change the risk profile.

PURPOSE

Run a weekly review workflow that turns infringement outputs into clear triage notes, root-cause prompts, and “what to check next” steps.

WHEN TO USE

• “Do a weekly tacho and WTD compliance review for these drivers.”
• “Triage these infringements and tell me what to check next.”
• “Summarise this PDF infringement report into actions and driver follow-ups.”

DO NOT USE WHEN…

• You only need a single driver-facing message (use the infringement coach skill).
• You want general rule explanations without records.

INPUTS

• REQUIRED:
  • Driver list + date range
  • Infringement summaries (CSV/PDF extract or pasted lines)
• OPTIONAL:
  • Any known operational context (routes, delays, multi-drop, ferry/train, staffing)
  • Prior RAG history
• EXAMPLES:
  • “Drivers A–F, last week’s infringement PDF attached; need weekly pack.”

OUTPUTS

• weekly-tacho-wtd-review.md (manager-facing)
• triage-actions-by-driver.md
• Success criteria:
  • Clear “next checks” per infringement type
  • Flags where investigation/discipline thresholds may be approaching (per your policy)

WORKFLOW

1. Confirm period and driver list.
   • IF missing → STOP AND ASK THE USER.
2. Normalise infringements into a per-driver list (facts only).
3. Pattern scan using references/common-infringement-patterns.md.
4. For each driver, generate:
   • Likely root-cause prompts (questions to ask, operational checks)
   • “What to check next” steps using assets/what-to-check-next-playbook.md
5. Produce weekly pack using assets/weekly-review-pack-template.md.
6. If RAG escalation depends on missing history → STOP AND ASK THE USER for counts/outcomes.

OUTPUT FORMAT

# triage-actions-by-driver.md
Period:
Sources:

## Driver [X]
Infringements (facts):
- …

What to check next:
- …

Root-cause prompts:
- …

Proposed follow-up:
- Coaching / monitoring / investigation trigger (per policy)

SAFETY & EDGE CASES

• If records appear incomplete (missing days, download gaps), flag as a risk/gap rather than guessing why.
• Avoid blame language; keep triage neutral.

EXAMPLES

• Input: “Weekly review for 12 drivers”
  • Output: weekly pack + per-driver triage actions and check-next prompts