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/elct9620/ai-coding-skills/inspectgit clone --depth 1 https://github.com/elct9620/ai-coding-skillsWrote 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/elct9620/ai-coding-skills/inspect)<a href="https://agentmods.dev/commands/elct9620/ai-coding-skills/inspect"><img src="https://agentmods.dev/badge/commands/elct9620/ai-coding-skills/inspect.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 | $0.00085 | $0.04445 |
| Opus 5 | $0.00043 | $0.02223 |
| Sonnet 5 | $0.00017 | $0.00889 |
| Haiku 4.5 | $0.00009 | $0.00445 |
Grade A, and why
inspect 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 5d 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 — 168 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Rule
The <execute name="main">ARGUMENTS</execute> is entry point for this command.
How Wide, How Deep
You are preparing for something the user has named — what they have noticed, or what they mean to take on next. Being prepared means knowing the whole this work sits in and what still stands between it and their goal. This runs before the work starts and the clarification that follows keeps filling it in, so a pass that stops early and says where it stopped serves better than one that tries to be exhaustive.
Width and depth cost different things, so they answer to different rules. Width is cheap — how the project got here from its log, what stands where from a tree, what else names this thing from a search — and it is what tells you which parts bear on the goal at all. Take it first and take it for the whole, not only for the place the target names: a part judged without the whole is judged without knowing what else moves when it moves. A narrow pass does not find less, it finds the same volume with nothing to rank it by.
Depth is expensive, so it is spent where width has already pointed. What licenses reading a file closely is a question it answers that a search cannot, so name the question first — is this how the codebase everywhere else acquires a collaborator? — and put it to whatever search the environment offers; a file you cannot attach a question to stays closed, whatever its size. When the file is long, search inside it and read the range that answers. Stop when one more file would no longer change what the work list holds or the order it should be taken in.
What bears on the goal is the one test both reading and reporting answer to — the goal being what the user wants to reach, not the state of the system around it. Something standing between them and it belongs to this round whether or not it sits where they happened to be looking; something true, real, and beside the point stays out, because a round that carries it has to rank it, and ranking it is what turns a short list into a survey.
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.
- 5d ago First seen · 168 lines · 0 tokens per session scan A d951f46f55d0
inspect is a command published in the GitHub repository elct9620/ai-coding-skills (5 stars, last pushed 9d ago), licensed Apache-2.0. It adds 85 tokens to every session and 4,445 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 commands, from other repositories
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.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.