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/panyam/mcpkit/example-upgradegit clone --depth 1 https://github.com/panyam/mcpkitWhat 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 | $0.02178 |
| Opus 5 | $0.00000 | $0.01089 |
| Sonnet 5 | $0.00000 | $0.00436 |
| Haiku 4.5 | $0.00000 | $0.00218 |
Grade A, and why
example-upgrade 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.
How it starts
The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Apply fixes that bring an mcpkit example into compliance with examples/CONVENTIONS.md. The action half of the audit/upgrade pair.
Args
$1— example name (e.g.discord-events) or relative path (e.g.examples/events/discord). If missing, ask.
Steps
-
Read
examples/CONVENTIONS.md. It's the spec; every fix must produce code that passes its checklist. -
Detect findings source — pipeline mode vs cold mode:
- Pipeline mode: scan recent conversation context for an
Audit complete: N pass, M failline emitted by/example-auditagainst$1. If found AND the audit is for the same target, parse the[FAIL]blocks (their stable IDs andWhere:hints) and skip to step 4. - Cold mode: no recent audit found. Run the audit checks internally (re-using the logic from
/example-audit— read CONVENTIONS.md §8, walk the checklist, gather findings). Don't emit the full audit report; just summarize:<N> issues found in examples/<path>: <id>, <id>, <id>.
- Pipeline mode: scan recent conversation context for an
-
If
build-brokenis among the findings, stop. Surface the build error to the user and ask them to fix it first (the upgrade skill can't safely apply other fixes against code that doesn't compile). Exit. -
Group findings by fix complexity and present them as a plan:
- Mechanical (single-edit, low-risk):
makefile-default-goal,filterargs-promoted,tui-helper,mode-helpers,logger-colorlogger,serve-srv-listenandserve,mux-withmux(when the side-endpoint already exists in a hand-rolled mux),walkthrough-md-fresh(re-runjust readme). - Structural (touches multiple files or rewrites a section):
dispatch-loop,makefile-baseline(full justfile rewrite),readme-quickstart/readme-what-it-demonstrates/readme-where-to-look(README sections),ui-extension,ui-bridge-template,ui-typed-app-tool. - Manual-only (skill flags but doesn't auto-apply):
client-close(placement depends on walkthrough flow),pretty-print-raw(depends on the step's intent),ui-readme-diagrams(needs human-authored sequence diagrams). Report these as "would need manual edits at :" and skip them.
- Mechanical (single-edit, low-risk):
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.
- yesterday First seen · 88 lines · 0 tokens per session scan A 66a8138a70aa
example-upgrade is a command published in the GitHub repository panyam/mcpkit (5 stars, last pushed 2d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 2,178 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
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.