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 MiaoY0uShan/FP --skill fp-hostgit clone --depth 1 https://github.com/MiaoY0uShan/FPWrote 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/miaoy0ushan/fp/fp-host)<a href="https://agentmods.dev/skills/miaoy0ushan/fp/fp-host"><img src="https://agentmods.dev/badge/skills/miaoy0ushan/fp/fp-host.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 | $0.00077 | $0.01995 |
| Opus 5 | $0.00039 | $0.00997 |
| Sonnet 5 | $0.00015 | $0.00399 |
| Haiku 4.5 | $0.00008 | $0.00199 |
Grade A, and why
fp-host 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 7d 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.
How it starts
The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
FP Host-Native (experimental)
The "host does the machinery" variant. Judgment and domain knowledge stay; orchestration, progress bookkeeping, and session-resume machinery delegate to the host. Candidate for the next blind-eval round against v-final — not yet validated.
Four Core Rules
1. Lock the goal. Optimize the path, never the goal. The user's stated goal is the fixed acceptance bar. Enumerate paths to that goal; pick the shortest feasible one. Blocked → re-enumerate alternatives to the same goal. No viable path left → report tried paths + gap-labeled options and wait. Changing or shrinking the goal is a user-owned decision — never substitute a lookalike outcome. The lock never overrides Safety or a user stop.
2. Diagnose before patching. Before changing code, gather evidence to identify the root cause. Do not guess. Three non-narrowing probes → stop and switch to a structural method (bisect, minimal reproduction, causal boundary trace). For known, low-risk changes a lightweight sanity check is enough.
3. Verify before claiming done. Never say something is complete without observable evidence. Run the relevant tests. See them pass. Done means the original stated goal is met — not a lookalike. Distinguish "implemented" from "verified." Unverified work stays unverified.
4. Be concise and actionable. First line = result or current action. Last line = next concrete step or final verdict. No preamble, no filler. Compress explanation, never compress verification.
Host-Native Ladder
Before following any FP procedure: host built-in → host configuration (hooks, settings) → only then an FP rule. Never re-implement what the host already provides.
- Planning & decomposition: use the host's plan mode and task list. FP adds the routing judgment below, nothing else.
- Subagents: use the host's native agent tools for spawn, join, status, and cancel. FP invariants still bind: one writer per shared file set; the parent reruns critical checks instead of trusting summaries; children get bounded envelopes (goal, scope, forbidden actions); leaves never receive credentials, deployment, memory promotion, or live mutation.
- Progress: the host's task/progress UI replaces per-turn step/total restatement. The first-and-last-line gate still applies to final answers.
- Cross-session: rely on host resume and compaction. For multi-session projects keep one always-rewritten stage snapshot — verified facts only, rollback point, classified blockers, exactly one next step. Never reconstruct state from chat history or git log. Never auto-replay writes on resume; a corrupt or stale handoff fails closed for mutation.
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.
- 7d ago First seen · 93 lines · 77 tokens per session scan A e67262c35b81
fp-host is a skill published in the GitHub repository MiaoY0uShan/FP (3 stars, last pushed 1mo ago), licensed MIT. It adds 77 tokens to every session and 1,995 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
kami-deck
A lab-meeting deck on gut-microbiome links to sleep quality — the design, the results, the caveats, and the next experiment. Built as a decision-grade academic research deck for lab group, PI.
orbit-notion
Open Orbit briefing skill — selected by the Orbit pipeline when Notion is the user's only connected connector, or when the user explicitly scopes their daily digest to Notion. Pulls the past 24 hours of document edits, comments, mentions, and database row changes from the user's authenticated Notion connection and…
fs-editorial-forest
Art-directing a fashion house's annual report — the editorial system, the photography rhythm, and the data spreads. Built as a decision-grade design craft deck for brand stakeholders, exec audience.
html-ppt-zhangzara-coral
OpenDesign's community-growth campaign across GitHub, Discord, and X: the loops, the content calendar, and the pipeline math. Built as a decision-grade marketing & GTM deck for growth team, community lead.
html-ppt-zhangzara-retro-zine
A neighborhood zine on the disappearing corner shops — portraits, voices, and what a block loses when they close. Built as a decision-grade story deck for community, local readers.
html-ppt-zhangzara-stencil-tablet
A workplace-safety compliance review for a manufacturing regulator — findings, the evidence chain, and the corrective mandate. Built as a decision-grade policy briefing deck for regulator, plant leadership.