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 commands/sergei-aronsen/claude-code-toolkit/audit-skipgit clone --depth 1 https://github.com/sergei-aronsen/claude-code-toolkitWrote 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/commands/sergei-aronsen/claude-code-toolkit/audit-skip)<a href="https://agentmods.dev/commands/sergei-aronsen/claude-code-toolkit/audit-skip"><img src="https://agentmods.dev/badge/commands/sergei-aronsen/claude-code-toolkit/audit-skip.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 | $0.00000 | $0.02454 |
| Opus 5 | $0.00000 | $0.01227 |
| Sonnet 5 | $0.00000 | $0.00491 |
| Haiku 4.5 | $0.00000 | $0.00245 |
Grade A, and why
audit-skip 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 4d 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 — 242 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/audit-skip — Append a False-Positive Exception to the Allowlist
Purpose
Append a structured exception entry to .claude/rules/audit-exceptions.md after the user
has confirmed a finding is not exploitable. Future /audit runs consult this file in
Phase 0 and skip matching findings.
Validation is hard-refusal — there is no --force flag. An exception against an untracked
path or an out-of-range line is a moving target the Council cannot reason about.
Usage
/audit-skip <file:line> <rule> <reason...>
/audit-skip <file:line> <rule> --council=council_confirmed_fp <reason...>
Examples:
/audit-skip src/foo.ts:42 SEC-XSS user input is escaped upstream by escapeHtml/audit-skip scripts/setup-security.sh:142 SEC-RAW-EXEC bash -c invocation runs hardcoded install commands, no user input/audit-skip api/users.ts:88 SEC-SQL --council=council_confirmed_fp prepared statement, scanner false-positive on string concat
When to Use
- After reviewing a
/auditfinding and confirming it is NOT exploitable in this codebase. - After the Council (
/council audit-review) returnsFALSE_POSITIVEand prompts you to persist the exception (use--council=council_confirmed_fpin that case per D-09). - DO NOT use to suppress findings you haven't read. The
Reasonfield is your justification — vague reasons like "false positive" or "not real" are inadequate.
Process
Step 1 — Parse Arguments
Parse all tokens after the command. Require at least three positional tokens (file:line, rule,
and at least one reason word). Extract the optional --council= flag from the reason-token
stream.
set -euo pipefail
# All args after the command, in raw form
ARGS=("$@")
if [ "${#ARGS[@]}" -lt 3 ]; then
printf 'audit-skip: usage: /audit-skip <file:line> <rule> <reason...>\n' >&2
exit 2
fi
FILE_LINE="${ARGS[0]}"
RULE="${ARGS[1]}"
# Split file:line on the LAST colon so paths containing : still work.
PATH_PART="${FILE_LINE%:*}"
LINE_PART="${FILE_LINE##*:}"
# Line must be a positive integer.
case "$LINE_PART" in
''|*[!0-9]*)
printf 'audit-skip: <file:line> must end with :<positive integer>, got %q\n' "$FILE_LINE" >&2
exit 2
;;
esac
# Reason tokens = everything after the rule, with --council=... extracted.
COUNCIL="unreviewed"
REASON_TOKENS=()
for tok in "${ARGS[@]:2}"; do
case "$tok" in
--council=council_confirmed_fp)
COUNCIL="council_confirmed_fp"
;;
--council=*)
printf 'audit-skip: --council= only accepts council_confirmed_fp (got %q)\n' "$tok" >&2
exit 2
;;
*)
REASON_TOKENS+=("$tok")
;;
esac
done
if [ "${#REASON_TOKENS[@]}" -eq 0 ]; then
printf 'audit-skip: reason is required (every token after the rule is joined as Reason)\n' >&2
exit 2
fi
# Join reason tokens with single spaces.
# Safety: always use printf '%s' to interpolate REASON — never echo (avoids
# backslash-escape interpretation on macOS).
REASON="$(printf '%s ' "${REASON_TOKENS[@]}")"
REASON="${REASON% }" # strip trailing space
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.
- 4d ago First seen · 242 lines · 0 tokens per session scan A 7a3d07c90b57
audit-skip is a command published in the GitHub repository sergei-aronsen/claude-code-toolkit (5 stars, last pushed 18d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,454 tokens. 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.
Other commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.