static-scan

A static code scanner for finding possible attack paths in an AI agent's source code. Its results are candidates for investigation, not proof that a vulnerability exists.

In plain words
What is it for?
Use it to scan an agent's source, classify candidate paths as RT-1 through RT-9 or RT-GUARD, and choose which findings to test with a red-team command.
Why use it?
It quickly highlights code paths that may deserve security testing without treating every warning as an exploitable flaw. A separate dynamic test is needed to confirm whether an attack works.

Command

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 commands/william2333zz/trustshell/static-scan
Clone the repo
git clone --depth 1 https://github.com/William2333ZZ/trustshell
Per session 32 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 298 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 $0.00032 $0.00298
Opus 5 $0.00016 $0.00149
Sonnet 5 $0.00006 $0.00060
Haiku 4.5 $0.00003 $0.00030

Measured yesterday against content hash def3277bbd1c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

static-scan 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 yesterday.

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.

commands/static-scan.md · 22 lines

What it actually says

Run TrustShell's static engine over the target agent's source and interpret the results.

Target source directory: $ARGUMENTS (if empty, ask the user which agent source to scan).

Steps:

  1. From the plugin root, run the static engine on the target: python3 "${CLAUDE_PLUGIN_ROOT}/scanner/static_scan.py" --source "$ARGUMENTS" --json
  2. Parse the JSON. For each finding, report: the RT class (RT-1..RT-9 / RT-GUARD), the file and symbol, and one line on why it's a candidate path.
  3. Be honest about what this is: triage, not a verdict. A flagged path is a CANDIDATE — the static engine says "attack here," it does NOT say "this is exploitable." Say so explicitly.
  4. Recommend the highest-value candidates to confirm dynamically, and tell the user they can run /red-team to attack the running agent and prove which candidates are real.

Do not overstate. Never call a candidate a vulnerability until an exploit confirms it.

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. yesterday First seen · 22 lines · 32 tokens per session scan A def3277bbd1c

Subscribe to this mod's changes

static-scan is a command published in the GitHub repository William2333ZZ/trustshell (1 stars, last pushed 1mo ago), licensed MIT. It adds 32 tokens to every session and 298 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 2026-08-31.