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 jdpalumbo2/workbench --skill clodex-auditgit clone --depth 1 https://github.com/jdpalumbo2/workbenchWrote 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/jdpalumbo2/workbench/clodex-audit)<a href="https://agentmods.dev/skills/jdpalumbo2/workbench/clodex-audit"><img src="https://agentmods.dev/badge/skills/jdpalumbo2/workbench/clodex-audit/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/jdpalumbo2/workbench/clodex-audit"><img src="https://agentmods.dev/badge/skills/jdpalumbo2/workbench/clodex-audit.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 | $0.00056 | $0.03237 |
| Opus 5.5 | $0.00022 | $0.01295 |
| Sonnet 5.5 | $0.00011 | $0.00647 |
| Haiku 4.5 | $0.00006 | $0.00324 |
Grade A, and why
clodex-audit 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 24d 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 — 280 lines — stays where its author put it; the contents beside it link to each section on GitHub.
clodex-audit — investigate, tag every claim, route the follow-ons
Overview
This lane earns its existence from usage evidence: two major audit asks ran manually in one weekend, both excellent, and their report shape is encoded here. It owes the run exactly three things:
- A report file in the repo, at the profile's plans directory (or a
reports/sibling when the repo has one), in the §5 shape. - Every claim in it tagged —
VERIFIED (<method>)orHYPOTHESIS (<what would confirm it>)— with the load-bearing ones also recorded as findings (finding:recorded,source: "audit") and the artifacts that prove them as evidence (verification:evidence). - The run closed with
release.stateuntouched atnot-started— an audit releases nothing, and its manifest must say so without ceremony.
The manifest's stage stays open for the life of an audit. The lane's
operational shape — open → investigate → report → closed — lives in this
document's sections, not in stage events: the vocabulary is frozen at 23
names, stage:plan:entered would be a lie, and a run at stage open with
lane audit is exactly how the router and the runs index recognize an
audit in flight. No build, no verify, no ship, ever; the follow-on work an
audit surfaces is routed per item (§5) into new runs.
You arrive here from clodex, which owns preflight, the profile, and the
run directory. Invoked without an absolute run dir → stop and invoke
clodex.
0. Paths and commands
CLODEX_HOME="${CLODEX_HOME:-$HOME/.claude/skills/clodex}"
STATE="$CLODEX_HOME/state/clodex_state.py"
RUN_DIR="<the absolute run dir the router handed you>"
SNAP="$(python3 "$STATE" rebuild "$RUN_DIR")"
REPO="$(printf '%s' "$SNAP" | python3 -c 'import json,sys;print(json.load(sys.stdin)["repo"])')"
cd "$REPO"
PROFILE="$REPO/.clodex/profile.json"
Engine verbs, exit codes, and the lock rules are the clodex skill's
§"Paths and commands" and §2. Events are written to files and piped on
stdin, or passed inline with append -e for small ones.
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.
- 24d ago Changed · +43 lines 5e24e6fe8ece
- 1mo ago First seen · 237 lines · 56 tokens per session scan A fe13fcd50432
clodex-audit is a skill published in the GitHub repository jdpalumbo2/workbench (1 stars, last pushed 15d ago), licensed MIT. It adds 56 tokens to every session and 3,237 once invoked, about $0.0002 per session on Opus 5.5. 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.
Other skills, from other repositories
code-that-fits-in-your-head
Software-engineering heuristics based on Mark Seemann's Code That Fits in Your Head (2021), updated for agent-driven development. Use when writing or reviewing code, refactoring accidental complexity or a Big Ball of Mud, controlling technical or architectural debt in generated code, designing APIs and invariants…
create-pr
Create pull requests following Sentry conventions. Use when opening PRs, writing PR descriptions, or preparing changes for review. Follows Sentry's code review guidelines.
git-pr-review
A read-only reviewer for GitHub pull requests, which are proposed code changes submitted for review. It produces an evidence-based report about whether a pull request should be merged.
code-remediate
Collect and resolve selected findings in one pass, then verify; bare PR targets use online items, while PR +review adds the latest matching artifact.
codex-review
Professional code review with auto CHANGELOG generation, integrated with Codex AI.
challenge-resolve
Independently review and fix a bounded task through convergence rounds, with evidence-backed closure and explicit stop/remediation decisions. Use only when the user explicitly requests an adversarial review-and-fix loop over an existing implementation, feature, problem, or diff; not for ordinary review (code-review)…