agentleFS
giorgosngiorgosn/diavgeia-watchdog/skills/detect-red-flags/SKILL.md

detect-red-flags

Skill2 starsChanged 5 months agoMITView on GitHub

Use when the user wants to find procurement red flags — contract-splitting (κατάτμηση), repeat/fragmented direct awards, suspicious clusters — in the pulled Greek data. Triggers on "find splitting", "red flags", "suspicious", "scan for", "who's gaming the threshold".

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run the deterministic detectors over the pulled data (data/contracts.jsonl, or data/awards.jsonl).

    line 8
  2. Distinctive:

    Always show the user the evidence links and tell them to open the source documents and seek right-of-reply before publishing anything.

    line 35

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Detect procurement red flags

Run the deterministic detectors over the pulled data (data/contracts.jsonl, or data/awards.jsonl). There is no AI in the verdict — it's set/threshold/date math, so the findings are reproducible and defensible.

node lib/detect.mjs splitting --in=data/contracts.jsonl --limit=30      # sub-threshold splitting
node lib/detect.mjs fragmentation --in=data/contracts.jsonl --limit=30  # repeat direct awards
node lib/detect.mjs splitting --json=leads.json                          # also write JSON for further work

What each detector means

  • splitting — same (authority × supplier × CPV), cumulative ≥ €30,000 within 365 days, while every individual award stays under €30,000. The classic "break it up to dodge the open tender" pattern.
  • fragmentation — same (authority × supplier), ≥ 4 direct awards within 90 days. Works even without amounts.

Each lead has a suspicion score (0–100) and concrete reasons (e.g. "4 of 6 awards just under the threshold", "all signed by X", "10 awards on the same day"), plus per-award evidence links to the public source document.

CRITICAL — how to present findings

  • These are LEADS for investigation, NOT accusations. Say so, every time.
  • High suspicion ≠ wrongdoing. Many high scorers are legitimate: hospital medical supplies, fuel, framework contracts, genuine emergencies, distinct lots. Read docs/METHODOLOGY.md.
  • Always show the user the evidence links and tell them to open the source documents and seek right-of-reply before publishing anything.
  • Surface the connections too (point them at the investigate skill): the same signer across many suppliers, or a supplier winning from many authorities, is the real story.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.