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 additive-agent-tool-surface-constructiongit 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/additive-agent-tool-surface-construction)<a href="https://agentmods.dev/skills/ychampion/cskill-agents/additive-agent-tool-surface-construction"><img src="https://agentmods.dev/badge/skills/ychampion/cskill-agents/additive-agent-tool-surface-construction.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 5.1 | $0.00034 | $0.00387 |
| Opus 5 | $0.00017 | $0.00193 |
| Sonnet 5 | $0.00007 | $0.00077 |
| Haiku 4.5 | $0.00003 | $0.00039 |
Grade A, and why
additive-agent-tool-surface-construction 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: Additive Agent Tool Surface Construction
Domain: tool-surface Trigger: Use this when bringing tools from newly connected MCP servers into the agent’s tool pool while preserving the parent context. Source Pattern: Distilled from reviewed subagent orchestration, isolation, and lifecycle implementations.
Core Method
When a child agent connects extra integrations, treat those capabilities as additions layered over the parent surface rather than as a replacement. Keep the parent client handles intact, append the agent-owned clients, and merge the newly discovered tool definitions alongside the inherited tool pool. Pass the merged clients, merged tools, and the child-owned cleanup handle downstream as one package so later stages can use the expanded surface without mutating the parent session.
Key Rules
- Always start with parent clients and append agent clients in order to avoid losing existing handles that the session still needs.
- Maintain agent tools separately so the agent can union them with the parent’s command/skill catalog without overwriting it.
- Return both the merged clients and tools along with the cleanup closure so callers can pass them downstream transparently.
- Log the before/after counts to aid debugging when tools disappear unexpectedly.
Example Application
When a sub-agent launches with extra MCP servers, use this pattern so it keeps the parent integrations it inherited while also gaining the newly connected agent-specific tools.
Anti-Patterns (What NOT to do)
- Don’t replace the parent client list with just the agent’s clients; that loses the original MCP endpoints.
- Don’t forget to pass the merged tools through the helper that runs the forked agent, or the new MCP servers will appear to succeed but never expose their tools.
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 · 28 lines · 34 tokens per session scan A e319ac3f07d0
additive-agent-tool-surface-construction is a skill published in the GitHub repository ychampion/cskill-agents (36 stars, last pushed 5mo ago), licensed MIT. It adds 34 tokens to every session and 387 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 2026-08-30.
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.
opensrc
Fetch dependency source code to give AI agents deeper implementation context. Use when the agent needs to understand how a library works internally, read source code for a package, fetch implementation details for a dependency, or explore how an npm/PyPI/crates.io package is built. Triggers include "fetch source for"…
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.
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.