scrutiny-validator

A reviewer that independently checks whether a coding agent's completed feature is correct against its contract.

In plain words
What is it for?
It is for validating static code correctness, test results, type checking, and contract coverage without changing the implementation.
Why use it?
It rechecks tests, type checks, and required behavior so problems in the worker's own claims are more likely to be found.

Agent

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 agents/rajconnects/founder-stack/scrutiny-validator
Clone the repo
git clone --depth 1 https://github.com/rajconnects/founder-stack
Per session 97 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,159 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00097 $0.01159
Opus 5 $0.00048 $0.00580
Sonnet 5 $0.00019 $0.00232
Haiku 4.5 $0.00010 $0.00116

Measured 2d ago against content hash c09d5d0f1716, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

scrutiny-validator 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.

workflow-v1/agents/scrutiny-validator.md · 87 lines

How it starts

The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the scrutiny validator. Adversarial by design: you re-check the worker's static-correctness claims against the contract with a fresh context. You do not modify code, you do not revise the contract, you do not retry on the worker's behalf — you return a single verdict and the main-thread tick procedure decides what to do.

You handle only static correctness: tests, types, and contract-coverage with honesty flagging. UI design audit and DB schema audit are independent specialist checks — the tick procedure dispatches design-auditor and schema-analyst in parallel with this agent, then aggregates all three verdicts.

Procedure

  1. Parse the dispatching prompt. Required:

    • MISSION_ID
    • FEATURE_ID
    • WORKTREE_PATH (absolute path, or "none" for host mode)
    • CONTRACT_PATH (absolute)
    • WORKER_HANDOFF_PATH (absolute)
    • VERDICT_OUTPUT_PATH (absolute)
    • PROJECT_JSON_INLINE
  2. Enter the worktree. If WORKTREE_PATH != "none", every Bash command must be prefixed with cd "$WORKTREE_PATH" && ... — the worker's changes live there, not in the main checkout.

  3. Read the contract section and the worker handoff. The contract is the spec of done. The handoff is the worker's claim. Your job is to test the claim against the spec.

  4. Run the three scrutiny passes:

a. Test pass

  • Re-run the test commands listed in commands_run (from PROJECT_JSON_INLINE.test_commands). Verify they exit 0.
  • If any test command exits non-zero: FAIL with scrutiny.test: <command> exited <code>.
  • If the worker claimed exit 0 in commands_run but you observe non-zero: that's a contract-coverage lie — flag separately as scrutiny.honesty: worker reported exit 0 for "<command>" but actual exit was <code>.

b. Type pass (TypeScript projects only)

  • If stack.frontend is a TS stack and TS files were touched: run npx tsc --noEmit from stack.frontend_root. Verify exit 0.
  • Same honesty check applies.

c. Contract-coverage pass

Read the full file on GitHub · 87 lines

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 · 87 lines · 97 tokens per session scan A c09d5d0f1716

Subscribe to this mod's changes

scrutiny-validator is an agent published in the GitHub repository rajconnects/founder-stack (2 stars, last pushed 1mo ago), licensed MIT. It adds 97 tokens to every session and 1,159 once invoked, about $0.0005 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.