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 cache-prefix-stability-for-subagentsgit 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/cache-prefix-stability-for-subagents)<a href="https://agentmods.dev/skills/ychampion/cskill-agents/cache-prefix-stability-for-subagents"><img src="https://agentmods.dev/badge/skills/ychampion/cskill-agents/cache-prefix-stability-for-subagents/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/cache-prefix-stability-for-subagents"><img src="https://agentmods.dev/badge/skills/ychampion/cskill-agents/cache-prefix-stability-for-subagents.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 | $0.00028 | $0.00408 |
| Opus 5.5 | $0.00011 | $0.00163 |
| Sonnet 5.5 | $0.00006 | $0.00082 |
| Haiku 4.5 | $0.00003 | $0.00041 |
Grade A, and why
cache-prefix-stability-for-subagents 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 1mo 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: Cache Prefix Stability For Subagents
Domain: agentic-loop Trigger: Apply when a subagent should share the parent's cached prompt prefix instead of recomputing a divergent request shape. Source Pattern: Distilled from reviewed subagent orchestration, isolation, and lifecycle implementations.
Core Method
Treat prompt-cache compatibility as an explicit contract when building child agent options. Preserve the cache-critical fields that define the API request prefix, including the merged tool pool, thinking configuration, query source, and content-replacement state, instead of letting child defaults drift away from the parent request. This keeps forked or exact-tool subagents eligible for the same cached prefix while still allowing non-cache-sensitive options to vary.
Key Rules
- Only inherit cache-critical request fields when the child is intentionally sharing the parent's prefix, such as the use exact tools path.
- Keep the resolved tool list, thinking configuration, and content-replacement state aligned with the parent request so cache keys do not diverge accidentally.
- Preserve the recursive-fork guard inputs like query source when they are part of the cached request prefix contract.
- Do not inherit unrelated defaults blindly; share only the fields that stabilize the prefix and let other child-specific behavior vary intentionally.
Example Application
If a forked review agent must reuse the parent’s prompt cache, build its options from the parent’s tool list and thinking config instead of resetting them to child defaults; the child can then reuse cached request prefixes while still emitting its own downstream messages.
Anti-Patterns (What NOT to do)
- Do not disable thinking or rebuild the tool pool for a cache-sharing child unless you also accept losing cache alignment.
- Do not rely on accidental structural similarity between parent and child requests; explicitly carry forward the fields that define the cache prefix.
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.
- 1mo ago First seen · 28 lines · 28 tokens per session scan A 069733d0ce11
cache-prefix-stability-for-subagents is a skill published in the GitHub repository ychampion/cskill-agents (36 stars, last pushed 6mo ago), licensed MIT. It adds 28 tokens to every session and 408 once invoked, about $0.0001 per session on Opus 5.5. 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.
gh
GitHub CLI skill for interacting with GitHub via the gh command line tool. Use when Bub needs to (1) Create, view, or manage GitHub repositories, (2) Work with issues and pull requests, (3) Create and manage releases, (4) Run and monitor GitHub Actions workflows, (5) Create and manage gists, (6) Install and manage…
brain-ingest
The process for digesting a conversation, document, or research result, classifying it, and writing it down as brain content (a root-page update or a new/updated page) through the brain CLI.
wish
Deliver one decided task end to end — admit it, work it in one worktree, gate, independent review, bounded repair, a merge-ready PR — or plan a multi-group wish when it is bigger than one task.
gate-control
Canonical VideoStudio review authorization and state-transition policy. Use after any Gate B/C/Preview/D decision, post-gate revision, resumed approval, or exhausted visual-QA result across COMPOSE/AUTO/GENERATE/EDIT; maps explicit user authority and durable artifact state to one next action with ovs gate transition.…
report
Investigate a failure to its root cause with grounded evidence, hand the diagnosis to fix, and create a GitHub issue only when asked.