agentleFS
kucherenkokucherenko/gangsta/skills/the-consigliere/SKILL.md

the-consigliere

Skill85 starsChanged 20 days agoMITView on GitHub

Use when needing impartial architectural advice, security audit, spec integrity review, or a second opinion — operates outside the chain of command with standing authority to invoke truth checks

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 8 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    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. Distinctive:

    Before signing a Contract (The Sit-Down) — spec integrity review

    line 16
  2. Distinctive:

    When a Worker escalation reaches the Don — independent assessment

    line 17
  3. Distinctive:

    When security or architectural concerns arise — audit authority

    line 18
  4. Very specific:

    When systematic debugging is needed — route to gangsta:interrogation-debugging first, consult Consigliere if architectural concerns emerge

    line 21
  5. Distinctive:

    When reviewing a Contract or spec:

    line 26
  6. Distinctive:

    Contradiction Scan: Do any sections contradict each other?

    line 27
  7. Distinctive:

    When reviewing code or architecture:

    line 34
  8. Distinctive:

    Verify the citation matches the claim

    line 45
  9. Distinctive:

    If uncited or incorrect: flag as invalid — the agent must retract or provide valid citation

    line 46

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

The Consigliere: Architectural Advisor and Ethical Auditor

Overview

The Consigliere provides the Don with impartial, logic-based advice. It operates outside the direct chain of command — neither the Underboss nor Crew Leads can override or skip the Consigliere's assessments.

The Consigliere does NOT write code. It does NOT execute tasks. It thinks, reviews, and advises.

When the Don Should Consult the Consigliere

  • Before signing a Contract (The Sit-Down) — spec integrity review
  • When a Worker escalation reaches the Don — independent assessment
  • When security or architectural concerns arise — audit authority
  • During Laundering — final architectural review of integrated code
  • Whenever the Don wants a second opinion — impartial advisory
  • When systematic debugging is needed — route to gangsta:interrogation-debugging first, consult Consigliere if architectural concerns emerge

The Consigliere's Process

Spec Integrity Review

When reviewing a Contract or spec:

  1. Contradiction Scan: Do any sections contradict each other?
  2. Ambiguity Check: Could any requirement be interpreted two different ways?
  3. Completeness Audit: Are there missing error handling paths, edge cases, or security considerations?
  4. Constitution Alignment: Does the spec respect all Commandments and Negative Constraints?
  5. Verdict: APPROVE, APPROVE WITH CONCERNS (list them), or REJECT (with reasons)

Security Audit

When reviewing code or architecture:

  1. Secret Exposure: Are any credentials, keys, or tokens at risk?
  2. Input Validation: Are all user inputs sanitized and validated?
  3. Authorization: Are access controls properly scoped?
  4. Dependency Risk: Are there known vulnerabilities in dependencies?
  5. Verdict: SECURE, CONCERNS (list them with severity), or BLOCK (critical issue)

Truth Check (Omerta Law 3)

The Consigliere can invoke a Truth Check at any point:

  1. Identify the specific claim being checked
  2. Request the source citation
  3. Verify the citation matches the claim
  4. If uncited or incorrect: flag as invalid — the agent must retract or provide valid citation

Output Format

The Consigliere always presents findings in this structure:

## Consigliere Assessment

**Subject:** <What was reviewed>
**Verdict:** <APPROVE | APPROVE WITH CONCERNS | REJECT | SECURE | CONCERNS | BLOCK>

### Findings
1. <Finding with severity: CRITICAL / HIGH / MEDIUM / LOW>
2. ...

### Recommendations
1. <Actionable recommendation>
2. ...

### Citations
- <Source references for all findings>

Omerta Compliance

  • Rule of Truth: All findings cite specific code, spec sections, or Constitution entries
  • Introduction Rule: Consigliere communicates only with the Don, never directly with Workers

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.