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/alexmond/alexmskills/statsgit clone --depth 1 https://github.com/alexmond/alexmskillsWrote 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/alexmond/alexmskills/stats)<a href="https://agentmods.dev/commands/alexmond/alexmskills/stats"><img src="https://agentmods.dev/badge/commands/alexmond/alexmskills/stats.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 | $0.00017 | $0.00913 |
| Opus 5 | $0.00009 | $0.00456 |
| Sonnet 5 | $0.00003 | $0.00183 |
| Haiku 4.5 | $0.00002 | $0.00091 |
Grade B, and why
stats scanned grade B with 1 finding 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.
Asks the agent to reveal its instructionsmediumSystem prompt leakage
Directions to print, repeat or translate the system prompt extract configuration the operator did not intend to expose.
description: Show prompt-coach health metrics — fires, praises, corrections, mastery, active rules How it starts
The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/prompt-coach:stats
Read the coach's state files and log to produce a compact health dashboard.
What to do
- Read the global state at
~/.claude/prompt-coach/state.json:prompt_count(total analyzed globally)normalization_stats.prompts_with_corrections,.tokens_corrected_total,.top_corrections(highest 5 by count)rules[*]— for each:status,fires_total,clean_streak,graduated_at(if any)
- Read the current repo's state at
.claude/prompt-coach/state.json(if present):prompt_count(this repo)reactivated(per-repo rule reactivations)
- Read the current repo's log at
.claude/prompt-coach/log.md(if present):- Count
outcome=nudged:*/outcome=collaborator:*(rewrite events) - Count
outcome=capped:*(rewrites silenced by the fatigue cap, v0.41+) - Count
outcome=praised:*(praise events, broken out by kind:mastery,first-after-fire,variable-ratio) - Count
outcome=skipped:conversational(short-circuited) - Count
outcome=no-emit/outcome=ack:*(analyzed but nothing fired)
- Count
- Acceptance ledger (v0.42+) — run
python3 <plugin>/scripts/config.py acceptance(or--json acceptance) for the acceptance rate: how often rewrites are taken (accepted + edited) vs rejected, globally and per rule.editedcounts as a hit;blindrejects (too fast to have read) are excluded from the rate. Surface the overall rate + any rule flagged⚠ dormant-risk(low acceptance → will be precision-gated out of firing).
Present as
prompt-coach v<version> — <today's date>
Volume (global): <prompt_count>
Volume (this repo): <this_repo_prompt_count>
Emit rate (this repo):
nudged: N | praised: M (mastery: X, first-after-fire: Y, variable-ratio: Z)
skipped: K | no-emit: L
rate: (N+M) / (N+M+L) = P% substantive prompts got signal
Rules:
mastered (dormant): <list rule ids graduated>
active (up to 6): <list rule ids currently practicing>
most-fired top 5: <rule: N fires>
Acceptance (v0.42+):
accepted A · edited E · rejected R · blind B → rate P%
dormant-risk: <rules with low acceptance, if any>
Typo normalization (global):
<prompts_with_corrections> prompts had corrections (<tokens_corrected_total> tokens)
top 5 corrections: <original: count>
Config in effect (this repo):
enabled: <bool> collaborator_gate: <bool>
praise_ratio: <n>
typo_tolerance: <n>
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 · 76 lines · 17 tokens per session scan B c14a17a7804f
stats is a command published in the GitHub repository alexmond/alexmskills (6 stars, last pushed 4d ago), licensed MIT. It adds 17 tokens to every session and 913 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 1 finding (asks the agent to reveal its instructions). 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.