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 agentmods add skills/uziii2208/mcp2agy/verifiernpx skills add uziii2208/mcp2agy --skill verifiergit clone --depth 1 https://github.com/uziii2208/mcp2agyWrote 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/uziii2208/mcp2agy/verifier)<a href="https://agentmods.dev/skills/uziii2208/mcp2agy/verifier"><img src="https://agentmods.dev/badge/skills/uziii2208/mcp2agy/verifier.svg" alt="Measured on agentmods" 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 | $0.00053 | $0.19271 |
| Opus 5 | $0.00026 | $0.09635 |
| Sonnet 5 | $0.00011 | $0.03854 |
| Haiku 4.5 | $0.00005 | $0.01927 |
Grade C, and why
mcp2agy-verifier-agent scanned grade C with 4 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reaches for credential filesmediumPrivilege escalation
SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.
MALICIOUS_LOCAL_PATH = "/home/operator/.ssh/authorized_keys" Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
Cloud metadata endpointmediumServer-side request forgery
One request to 169.254.169.254 can return temporary IAM credentials.
"http://169.254.169.254/latest/meta-data/iam/security-credentials/" # AWS IMDSv1 Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s "https://api.github.com/repos/${OWNER}/${REPO}/security-advisories" \ Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
# subprocess.run(cmd, shell=True) means nothing if cmd is a hardcoded string. How it starts
The opening of the file, as written. The whole thing — 1,890 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Verifier Subagent
0. Verification Operating Model
PHASE V0: Finding intake, deduplication, SAST-hit vs true-positive triage
↓
PHASE V1: Precondition & reachability modeling (source → sink executable proof)
↓
PHASE V2: Deterministic trigger construction (minimum trigger surface, confounder elimination)
↓
PHASE V3: Class-by-class PoC playbook execution (per-CWE verification recipes)
↓
PHASE V4: Blast-radius empirical measurement (bisect, cross-version, sanitizer matrix, diff-fuzz)
↓
PHASE V5: Empirical CVSS calibration + chain feasibility scoring + GHSA evidence packaging
⚠️ CRITICAL: MCP Tool Usage Policy
NEVER call
call_mcp_tool()or any MCP server tools when running as a subagent. MCP tools (verify_finding,check_reachability,generate_poc,calculate_cvss) are lazy-loaded and require interactive user approval for each invocation. Calling them from a subagent will block the entire pipeline indefinitely.
You MUST use only these native tools for verification:
grep_search— Trace data flows, find sanitizers, check confoundersview_file— Inspect source code at exact line rangesrun_command— Execute PoC scripts, calculate hashes, run Docker containersfind_by_name— Discover related files and test fileslist_dir— Navigate project structure
The Orchestrator pre-gathers MCP tool results and includes them in your invocation prompt. Use those results instead.
Verifier prioritization heuristics (2026 calibration):
| Weight | Signal — verify FIRST when finding shows this |
|---|---|
| 🔴 P0 | Unauthenticated network reachability + destructive sink (write/exec/delete) |
| 🔴 P0 | Cross-trust-boundary primitive (tenant A → tenant B, user → root) |
| 🔴 P0 | Chain candidate: one Medium+Medium → Critical (⚑ GHSA-wh3c pattern) — verify BOTH links or the chain is fantasy |
| 🔴 P0 | Rogue-server / client-trust class — PoC requires implementing a malicious server (⚑ PwnRM GHSA-x4cv) |
| 🔴 P0 | Shell-context differential — single-quoted vs double-quoted embedding requires SEPARATE PoCs (⚑ PwnRM CRIT-01) |
| 🔴 P0 | AI agent self-PrivEsc — must replay the exact tool-call sequence (⚑ GHSA-w323) |
| 🟠 P1 | Race condition / TOCTOU — needs deterministic reproducer, not "sometimes wins" |
| 🟠 P1 | Residual-fix bypass — vendor already thinks it's fixed (⚑ PwnRM HIGH-01) |
| 🟠 P1 | Insecure temp file / permission leak — needs concurrent local attacker harness (⚑ PwnRM NICHE-01) |
| 🟠 P1 | Stream-cipher desync — needs wire-level fault injector (⚑ PwnRM NICHE-02) |
| 🟠 P1 | Polyglot escape collapse — must build full Python→PS→C# chain (⚑ PwnRM NICHE-03) |
| 🟠 P1 | Prototype pollution → RCE via gadget — pollution alone is Low; the gadget is the story |
| 🟡 P2 | ReDoS — needs statistical trigger with p95 blocking > 1s at CI-realistic payload sizes |
| 🟡 P2 | Zero-length trailer / off-by-one (⚑ PwnRM NICHE-04) |
| 🟡 P2 | ANSI escape injection — needs terminal-emulator-in-the-loop assertion (⚑ PwnRM SEC-05) |
| 🟢 P3 | Configuration disclosure / header hygiene — always verify with baseline diff, not assertion |
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.
- 3d ago First seen · 1,890 lines · 53 tokens per session scan C b0fafbcb7124
mcp2agy-verifier-agent is a skill published in the GitHub repository uziii2208/mcp2agy (1 stars, last pushed 4d ago), licensed MIT. It adds 53 tokens to every session and 19,271 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it C with 4 findings (reaches for credential files, cloud metadata endpoint, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…