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 rjmurillo/ai-agents --skill ai-agents-validation-and-qagit clone --depth 1 https://github.com/rjmurillo/ai-agentsWrote 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/rjmurillo/ai-agents/ai-agents-validation-and-qa)<a href="https://agentmods.dev/skills/rjmurillo/ai-agents/ai-agents-validation-and-qa"><img src="https://agentmods.dev/badge/skills/rjmurillo/ai-agents/ai-agents-validation-and-qa/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/rjmurillo/ai-agents/ai-agents-validation-and-qa"><img src="https://agentmods.dev/badge/skills/rjmurillo/ai-agents/ai-agents-validation-and-qa.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Agent Snooping · line 188 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
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 | $0.00126 | $0.04546 |
| Opus 5.5 | $0.00050 | $0.01818 |
| Sonnet 5.5 | $0.00025 | $0.00909 |
| Haiku 4.5 | $0.00013 | $0.00455 |
Grade C, and why
ai-agents-validation-and-qa scanned grade C with 1 finding 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 10d 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
Verified 2026-07-29 against the working tree (issue #3828 re-verification pass; the earlier 2026-07-03 pass had rotted for `pyproject.toml`, both `conftest.py` files, `.github/workflows/pytest.yml`, and `.claude/rules/ge How it starts
The opening of the file, as written. The whole thing — 210 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ai-agents Validation and QA
This repo runs verification-based governance: a claim is true when a gate, test, or command output says it is, not when an agent asserts it. This skill defines the evidence bar for code changes, where tests live and how they are collected, the anti-pattern canon (each entry paid for by a real incident), and the QA gate semantics at session end.
Audience: a zero-context contributor (human or model) about to write, run, or skip tests in this repository.
Triggers
what counts as evidencehow do I test this changerun skill testscan I skip QA
Scope Boundaries
| You want | Use instead |
|---|---|
| Triage a red CI check or a blocked push | ai-agents-debugging-playbook |
| Measure drift, budgets, telemetry, coverage trends | ai-agents-diagnostics-toolkit |
| Probe an external tool's runtime behavior empirically | ai-agents-empirical-probe-toolkit |
| Understand which gates a change class triggers | ai-agents-change-control |
| Decide whether two pieces of evidence are independent, or whether a named check ran | evidence-provenance |
| Session-log mechanics beyond the QA row | .claude/rules/session-logs.md |
Process
Phase 1: Internalize the evidence bar
.agents/governance/TESTING-RIGOR.md is BLOCKING for code changes (its own Status line, TESTING-RIGOR.md:5). The bar, per function you add or change:
| Requirement | Concretely |
|---|---|
| Positive test | Valid input produces the expected output |
| Negative test | Invalid input produces the language's idiomatic error, asserted |
| Edge tests | Whitespace, empty, None, wrong type |
| Every error branch | Each raise / error-return path exercised |
| Every conditional branch | Including branches that only change user-facing strings |
| Mocked I/O | No live subprocess, API, or file dependencies in unit tests |
| CLI contract | argv-failure exits, exit codes, stdout vs --output tested |
| Coverage proof | 100% block coverage on changed files (see Phase 4) |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago Changed 4897222beff7
- 12d ago Changed · +2 lines 63daad899538
- 14d ago Changed · +1 lines 3ee348bf8f2f
- 16d ago Changed 73c9ec07931f
- 18d ago Changed 16fd9264eb62
- 26d ago Changed · -2 lines scan D → C a642a4e896e4
- 1mo ago Changed · +6 lines ceef379b79ef
- 1mo ago Changed 4f808e34a328
- 1mo ago Changed · +10 lines scan B → D cfa183299ef6
- 1mo ago First seen · 193 lines · 126 tokens per session scan B 7f4849e7e18a
ai-agents-validation-and-qa is a skill published in the GitHub repository rjmurillo/ai-agents (49 stars, last pushed today), licensed MIT. It adds 126 tokens to every session and 4,546 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
orchestrated-execution
Execute work units through the rigorous 4-phase Metaswarm cycle (Implement -> Validate -> Adversarial Review -> Commit) with independent quality gate enforcement.
Pair Programming
AI-assisted pair programming with multiple modes (driver$navigator$switch), real-time verification, quality monitoring, and comprehensive testing. Supports TDD, debugging, refactoring, and learning sessions. Features automatic role switching, continuous code review, security scanning, and performance optimization with…
phoenix-test
Drive development and bug-fixing with tests where the test IS the objective phoenixsense gate. Write a failing test before the code, reproduce a bug with a test before fixing it, and let phoenixsense decide pass/fail — never "looks right". Use when implementing logic, fixing a bug, changing behavior, or when the user…
phoenix-review
Review completed work against the Intent Contract's acceptance criteria using objective evidence, not opinion — re-run every check, confirm no regressions, inspect the tamper-evident trace, and surface only real issues (bugs, unmet criteria, regressions). Use before shipping, after a feature is built, or when the user…
parity-guard-test-pattern
When checking that all callers of a contract implement it correctly, you get false positives from wrapper delegators.
team-implement
Use for executing implementation plans only on explicit request. Never infer from a ready plan.