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 ychampion/cskill-agents --skill mcp-skill-vs-prompt-discovery-filteringgit clone --depth 1 https://github.com/ychampion/cskill-agentsWrote 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/ychampion/cskill-agents/mcp-skill-vs-prompt-discovery-filtering)<a href="https://agentmods.dev/skills/ychampion/cskill-agents/mcp-skill-vs-prompt-discovery-filtering"><img src="https://agentmods.dev/badge/skills/ychampion/cskill-agents/mcp-skill-vs-prompt-discovery-filtering/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/ychampion/cskill-agents/mcp-skill-vs-prompt-discovery-filtering"><img src="https://agentmods.dev/badge/skills/ychampion/cskill-agents/mcp-skill-vs-prompt-discovery-filtering.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 | $0.00036 | $0.00423 |
| Opus 5 | $0.00018 | $0.00211 |
| Sonnet 5 | $0.00007 | $0.00085 |
| Haiku 4.5 | $0.00004 | $0.00042 |
Grade A, and why
mcp-skill-vs-prompt-discovery-filtering 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 8d 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.
What it actually says
SKILL: MCP Skill vs Prompt Discovery Filtering
Domain: tool-design Trigger: Apply when SkillTool enumerates commands to decide which slash commands should be available, ensuring MCP prompt skills are disambiguated from generic prompts. Source Pattern: Distilled from reviewed command discovery, MCP integration, and skill-surface implementations.
Core Method
Inspect the MCP-discovered command surface, keep only entries that are actually prompt-based skills, and merge them with the local command registry into one unified list. Deduplicate by command name before validation or execution so MCP-discovered prompts cannot shadow or duplicate local skills by accident. Preserve source metadata so permission checks, telemetry, and UX can still tell whether a skill came from MCP or from a local registry.
Key Rules
- Always filter MCP commands by
cmd.type === 'prompt'before merging; prompts of other types should stay hidden because SkillTool only handles prompt-based skills. - Merge the filtered MCP list with the local command registry and dedupe by name so remote skills do not conflict with bundled commands.
- Preserve metadata like
loadedFromor other source identifiers so permission checks and telemetry can tell whether the skill came from MCP or local registries.
Example Application
If an MCP server exposes a prompt named changelog and the local registry already has one, run this filter before permission checks so the merged skill surface stays coherent and the MCP entry does not silently shadow the local one.
Anti-Patterns (What NOT to do)
- Do not merge MCP prompts without filtering type or deduping, which would let MCP prompts masquerade as local skills and break strict permission rules.
- Do not treat MCP prompts as regular
promptcommands without markingloadedFrom:'mcp', because telemetry and permission prompts rely on knowing which commands were discovered versus bundled.
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.
- 8d ago First seen · 29 lines · 36 tokens per session scan A 8445ab68e9f8
mcp-skill-vs-prompt-discovery-filtering is a skill published in the GitHub repository ychampion/cskill-agents (36 stars, last pushed 5mo ago), licensed MIT. It adds 36 tokens to every session and 423 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
grape
Use Grape MCP for Codex context continuity in coding repositories. Use when a task needs repeated-turn context, omitted context restore, stale-context checks, invalidation checks, or safe continuity across branch and dirty-worktree changes.
review
Validate plans, execution, or PRs against wish criteria — returns SHIP / FIX-FIRST / BLOCKED with severity-tagged gaps.
work
Execute an approved wish plan — orchestrate subagents per task group with fix loops, validation, and review handoff.
brainstorm
Explore ambiguous or early-stage ideas interactively — tracks wish-readiness and crystallizes into a design for wish.
fix
Dispatch fix subagent for FIX-FIRST gaps from review, re-review, then diagnose unresolved failures after 2 loops.
genie
Entry point for Genie operations — routes bug reports, questions, and operational commands, resumes existing lifecycle state, and orchestrates work that needs durable planning or coordination. Other ordinary requests bypass the lifecycle with a one-line notice unless the user asks for Genie.