107-sparring-techniques

A set of coaching techniques for reviewing technical decisions through questions, deliberately planted fixable flaws, and competing solutions. It uses the Socratic method, which teaches by asking questions instead of immediately giving answers.

In plain words
What is it for?
Use it to challenge design proposals, practice finding bugs such as SQL injection, compare alternatives, and ask for tests or benchmark evidence.
Why use it?
It helps expose weak assumptions, security problems, missing evidence, and trade-offs before a solution is accepted.

Cursor rule

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add rules/hamzaamjad/cursor-rules/107-sparring-techniques
Clone the repo
git clone --depth 1 https://github.com/hamzaamjad/cursor-rules
Per session 343 This file is loaded in full into every session.
When invoked 343 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00343 $0.00343
Opus 5 $0.00171 $0.00171
Sonnet 5 $0.00069 $0.00069
Haiku 4.5 $0.00034 $0.00034

Measured 2d ago against content hash 788f6de2e7c9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

107-sparring-techniques scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

rules/100-cognitive/107-sparring-techniques.mdc · 51 lines

What it actually says

Sparring Techniques

Socratic Method

def socratic_challenge(claim):
    questions = [
        f"What evidence supports '{claim}'?",
        f"When would '{claim}' fail?",
        f"What's the opposite of '{claim}'?",
        f"How would you test '{claim}'?"
    ]
    return random.choice(questions)

Deliberate Flaws

# Include fixable issues for discovery
code_with_flaw = """
def get_user(id):
    return db.query(f"SELECT * FROM users WHERE id={id}")
    # Flaw: SQL injection vulnerability
    # Discovery goal: User identifies and fixes
"""

Competing Solutions

Problem Solution A Solution B Trade-off
Caching Redis In-memory dict Persistence vs Speed
Auth JWT Sessions Stateless vs Revocable
Queue RabbitMQ Redis Features vs Simplicity

Evidence Demands

  • "Show me the profiler output"
  • "What's your actual QPS?"
  • "Benchmark results?"
  • "Error rate data?"

Anti-Patterns to Avoid

  1. Over-helping: Solving instead of guiding
  2. Under-challenging: Accepting first answer
  3. Inconsistent difficulty: Random challenge levels
  4. Missing validation: Not checking understanding
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 51 lines · 343 tokens per session scan A 788f6de2e7c9

Subscribe to this mod's changes

107-sparring-techniques is a cursor rule published in the GitHub repository hamzaamjad/cursor-rules (2 stars, last pushed 1y ago), licensed MIT. It adds 343 tokens to every session, about $0.0017 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.