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.
npx skills add Vimalk0703/shipworthy --skill confidence-based-strictnessgit clone --depth 1 https://github.com/Vimalk0703/shipworthyWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/vimalk0703/shipworthy/confidence-based-strictness)<a href="https://agentmods.dev/skills/vimalk0703/shipworthy/confidence-based-strictness"><img src="https://agentmods.dev/badge/skills/vimalk0703/shipworthy/confidence-based-strictness/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/vimalk0703/shipworthy/confidence-based-strictness"><img src="https://agentmods.dev/badge/skills/vimalk0703/shipworthy/confidence-based-strictness.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00041 | $0.01467 |
| Opus 5 | $0.00020 | $0.00733 |
| Sonnet 5 | $0.00008 | $0.00293 |
| Haiku 4.5 | $0.00004 | $0.00147 |
Grade A, and why
confidence-based-strictness 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Confidence-Based Strictness
Core Rule
When confidence is low, verification must be high. The more uncertain the system is about the correctness of generated code, the more guardrails should activate and the stricter they should be.
Confidence Levels
Level 1: High Confidence (Routine)
Indicators:
- Well-known pattern (CRUD endpoint, form validation, standard middleware)
- Single clear approach — no ambiguity about the right solution
- Existing tests cover the area being modified
- Small change to an established codebase
- Language/framework the project already uses
Guardrail Response:
- Standard enforcement — normal quality gates
- Tests must pass, lint must pass
- No extra verification needed
Level 2: Moderate Confidence (Non-Trivial)
Indicators:
- Multiple valid approaches exist (design decision needed)
- Modifying code without existing test coverage
- New feature in an established codebase
- Performance-sensitive code where the "obvious" approach may not be optimal
- Cross-module changes that could have side effects
Guardrail Response:
- All standard guardrails plus:
- Require tests before and after (TDD flow)
- Run full test suite, not just affected tests
- Verify no unintended side effects on related modules
- Suggest code review before merge
Level 3: Low Confidence (High Uncertainty)
Indicators:
- Unfamiliar domain (cryptography, financial calculations, distributed consensus)
- Complex algorithm with edge cases
- No existing tests in the affected area
- Security-critical path (authentication, authorization, payment)
- Concurrency or race condition potential
- External API integration with unclear behavior
- Regulatory implications (HIPAA, PCI-DSS, GDPR)
Guardrail Response:
- All moderate guardrails plus:
- Mandatory threat model review (
shipworthy:threat-modeling) - Edge case enumeration — list and test boundary conditions explicitly
- Require verification evidence for every claim (see
shipworthy:verification-before-completion) - Suggest splitting the change into smaller, independently verifiable steps
- Activate domain-specific skills (compliance-awareness, bias-detection, etc.)
- Flag for human review with specific review focus areas
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.
- 7d ago First seen · 149 lines · 41 tokens per session scan A e123277089e7
confidence-based-strictness is a skill published in the GitHub repository Vimalk0703/shipworthy (7 stars, last pushed 5mo ago), licensed MIT. It adds 41 tokens to every session and 1,467 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
testing-strategy
Design test strategies and test plans with coverage targets. Complements /draft:coverage which measures what this skill plans. Auto-loaded by /draft:implement before TDD.
dare-execute
Executa uma task específica com implementação de código e testes. Use quando o usuário aprovar TASKS.md e quiser executar uma task. Implementa o código, roda testes (Ralph Loop) e valida até passar.
enterprise-forge
Adversarial code review with mechanical checks, contract probing, and 5 adversarial lenses. Bugs recycle to contract for full TDD treatment. 3-fail circuit breaker prevents infinite loops. Use after enterprise-review.
gentle-ai
Use Gentle AI harness discipline for Pi work: clarify first, preserve OpenSpec artifacts, use strict TDD where available, delegate through subagents when useful, and protect review workload.
plan-pipeline-execute
Execute a validated plan: worktree isolation, TDD scaffolding, level-based parallel agents, quality gate with smoke test, PR creation and merge. Handles everything through to merged PR.
coverage
Compute code coverage for active track or module. Targets 95%+ coverage with report and justification for uncovered lines. Complements TDD workflow.