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 skills/haabe/mycelium/metrics-pullnpx skills add haabe/mycelium --skill metrics-pullgit clone --depth 1 https://github.com/haabe/myceliumWrote 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/haabe/mycelium/metrics-pull)<a href="https://agentmods.dev/skills/haabe/mycelium/metrics-pull"><img src="https://agentmods.dev/badge/skills/haabe/mycelium/metrics-pull.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 | $0.00043 | $0.03656 |
| Opus 5 | $0.00022 | $0.01828 |
| Sonnet 5 | $0.00009 | $0.00731 |
| Haiku 4.5 | $0.00004 | $0.00366 |
Grade B, and why
metrics-pull scanned grade B with 2 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 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
For each source, find the most recent snapshot in `.claude/evals/metrics/<source>/` that is NOT today's. If none exists, skip delta computation for that source and note "first snapshot" in the report. Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- All three adapters' `gh api` / `curl` / `stripe` commands issued in one parallel batch. How it starts
The opening of the file, as written. The whole thing — 241 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Metrics Pull Skill
Unified pull across all metric sources configured in .claude/jit-tooling/active-metrics.yml. Dispatches to per-source adapters (starting with github), normalizes to a shared schema, computes deltas, and drafts canvas evidence entries.
This skill is the L0/L1/L2/L5 evidence-gathering loop. Replaces manual "I checked the dashboard" reports with timestamped, sourced, diffable data.
When to Use
- Before
/mycelium:diamond-assessat L0/L1/L2/L5, if the newest snapshot is >7 days old. - After any public activity (launch, post, conference mention) — typically 24-48h later to capture the bump.
- Weekly baseline, regardless of activity.
- Ad hoc when the user wants a fresh read.
Precondition: .claude/jit-tooling/active-metrics.yml exists
If .claude/jit-tooling/active-metrics.yml does NOT exist:
- Tell the user: "No metric sources are configured. Running
/mycelium:metrics-detectfirst." - Invoke
/mycelium:metrics-detect(or follow${CLAUDE_PLUGIN_ROOT}/jit-tooling/metrics-detector.md). - After detection completes, return to this skill.
Workflow
Step 1: Load configuration
Read .claude/jit-tooling/active-metrics.yml. Filter to sources with status: active. For each, verify the adapter file exists at ${CLAUDE_PLUGIN_ROOT}/jit-tooling/metrics-adapters/<source>.md. If missing, follow ${CLAUDE_PLUGIN_ROOT}/jit-tooling/metrics-adapters/GENERATING.md to generate it.
If confirmed_by_user: false, ask the user to confirm the source list before proceeding.
Step 2: Dispatch adapters in parallel
For each active source, follow its adapter's "Pull" and "Normalize" sections. Adapters are independent — run them in parallel (bash jobs, multiple tool calls in one message).
Per-source handling:
- Honor the adapter's
credential_requirement. If missing, skip the source and report to user; do not fail the entire pull. - Honor the adapter's rate-limit and retry rules.
- If the adapter returns
fetch_status: partial, include partial data but mark the snapshot.
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 · 241 lines · 43 tokens per session scan B a8c8f8c5ad9c
metrics-pull is a skill published in the GitHub repository haabe/mycelium (45 stars, last pushed today), licensed MIT. It adds 43 tokens to every session and 3,656 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
context-manager
Context management skill providing discovery, fetching, harvesting, extraction, compression, organization, cleanup, and guided workflows for project context.
tao-setup
One-time session setup and orchestration map for the TAO skill bank. Run this first when the TAO skills were installed individually (e.g. from a public skills catalog) so the session gets the cross-skill discovery flow, credential checks, and host preflight that the bundled plugin hook would otherwise inject…
tao-list-capabilities
Answer what the TAO Skill Bank plugin can do by generating the response from packaged application, data, model, AutoML, and platform manifests. Use when the user asks "what can TAO Skill Bank do", "list TAO models", "which TAO workflows are available", or "what supports AutoML".
find-skills
Discovers, searches, and installs skills from multiple AI agent skill marketplaces (400K+ skills) using the SkillKit CLI. Supports browsing official partner collections (Anthropic, Vercel, Supabase, Stripe, and more) and community repositories, searching by domain or technology, and installing specific skills from…
janitor-discover
Find new skills on GitHub or check a specific skill before installing. Use when the user wants to search for skills, evaluate a skill URL, check overlap and security risk before installing, or compare a local skill against alternatives. Trigger with '/janitor-discover'.
gingiris-user-interview
🇺🇸 User Interview & Cold-Start Operations Playbook — Battle-tested framework from HeyGen's 937 interviews to PMF. Complete SOP for user screening, interview execution, Beta testing methodology, cold-start user operations, churn analysis, and user grading systems. 🇨🇳 用户访谈与冷启动运营实战手册 — 基于 HeyGen 937 场访谈验证的 PMF…