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 escoffier-labs/skillet --skill expeditegit clone --depth 1 https://github.com/escoffier-labs/skilletWrote 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/escoffier-labs/skillet/expedite)<a href="https://agentmods.dev/skills/escoffier-labs/skillet/expedite"><img src="https://agentmods.dev/badge/skills/escoffier-labs/skillet/expedite/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/escoffier-labs/skillet/expedite"><img src="https://agentmods.dev/badge/skills/escoffier-labs/skillet/expedite.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 | $0.00078 | $0.01300 |
| Opus 5 | $0.00039 | $0.00650 |
| Sonnet 5 | $0.00016 | $0.00260 |
| Haiku 4.5 | $0.00008 | $0.00130 |
Grade A, and why
expedite 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 12d 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
expedite
The expeditor at the pass drives tickets from the board to plated, in priority order, and lets nothing leave half-cooked. This skill takes an audit backlog (from line-check, bug-hunt, security-sweep, latent-premises, or retry-safety) and executes it: highest-leverage finding first, one focused change at a time, each one verified before the next is started.
The audit roster is read-only on purpose. This is the separate step it hands off to. Findings already carry severity, effort, location, and a fix. The work is execution discipline, not re-analysis.
Preconditions
Check before touching anything; stop and report if any fail:
- A backlog exists. The input is an audit report following the report contract (a
## Backlogsection, or## Findingsto sort yourself). No report? Run line-check, bug-hunt, security-sweep, latent-premises, or retry-safety first, do not freelance. - Working on a branch, never the default branch. Create one if needed (
expedite/<audit>-<date>). - Clean tree to start, so each finding's change is isolable.
Execution
Work the backlog in leverage order (impact relative to effort), the order the report already sorted. If the report has only a ## Findings section and no ## Backlog, sort it yourself first: cheap high-impact items float to the top, severity breaks ties. An L-effort finding that cannot land in one focused session is not a half-job: attempt it only if it is cleanly isolable, otherwise surface it under Remaining with an estimate rather than leaving the tree mid-change.
For each finding:
- Read the finding, then the actual code at its "Where". Confirm it still reproduces. Reality drifts from the report, and earlier fixes in this run may have moved or resolved later findings; a finding already fixed gets marked done and skipped, not "fixed" again. If an earlier change relocated the code, update the "Where" before proceeding.
- Make the smallest change that resolves it. One finding, one focused change. Do not bundle, do not refactor adjacent code the finding did not name, do not gold-plate.
- For a bug or behavior change, write the failing test first (taste). Watch it fail, then make it pass; the failing test is the proof the change does what the finding claims. For a defect whose cause is unclear, reproduce it and trace it to the root cause before proposing a fix (refire). A bandaid that masks the symptom is not a fix.
- Verify the finding is actually gone. Re-check the "Where", run the relevant tests, run the command the fix claims to repair, and read the output before marking anything done (check). Evidence, not assertion.
- Commit per finding with a message naming the effect, not the plumbing. Keeps the diff reviewable and lets any single fix be reverted alone.
- Mark the finding done in the backlog with a one-line note (commit SHA or what changed), so the report becomes a live worklist.
What ships with it
2 files 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.
- 12d ago First seen · 78 lines · 78 tokens per session scan A 948de6698f81
expedite is a skill published in the GitHub repository escoffier-labs/skillet (4 stars, last pushed 11d ago), licensed MIT. It adds 78 tokens to every session and 1,300 once invoked, about $0.0004 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.
Other skills, from other repositories
craft-fix
The Craftsman standard for driving fixes against an existing craft-audit workspace — "fix the findings", "fix SEC-003", "work through the audit", "start the climb". An ACTION skill, not a domain: it picks findings off the master tracker's climb sequence, re-verifies each is still real, gets the user's approval, and…
production-audit
Exhaustive production-readiness audit of any codebase or product. Runs diverse-angle discovery passes until two consecutive passes find nothing new, adversarially verifies every finding against the real code, and reports one flat severity-tagged list with file:line evidence. No summary, no hedging.
spark-engineer
Use when writing Spark jobs, debugging performance issues, or configuring cluster settings for Apache Spark applications, distributed data processing pipelines, or big data workloads. Invoke to write DataFrame transformations, optimize Spark SQL queries, implement RDD pipelines, tune shuffle operations, configure…
debugging-wizard
Parses error messages, traces execution flow through stack traces, correlates log entries to identify failure points, and applies systematic hypothesis-driven methodology to isolate and resolve bugs. Use when investigating errors, analyzing stack traces, finding root causes of unexpected behavior, troubleshooting…
Reverse Engineering & Binary Analysis
Binary analysis, assembly interpretation, disassembly, decompilation, firmware RE, and protocol reverse engineering.
log-analyzer
A log-analysis helper that reads application and system logs to find unusual patterns and likely causes. Logs are records of events such as errors, requests, warnings, and service activity.