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 ProsusAI/prism --skill synthesize-decisionsgit clone --depth 1 https://github.com/ProsusAI/prismWrote 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/prosusai/prism/synthesize-decisions)<a href="https://agentmods.dev/skills/prosusai/prism/synthesize-decisions"><img src="https://agentmods.dev/badge/skills/prosusai/prism/synthesize-decisions/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/prosusai/prism/synthesize-decisions"><img src="https://agentmods.dev/badge/skills/prosusai/prism/synthesize-decisions.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 | $0.00097 | $0.02437 |
| Opus 5.5 | $0.00039 | $0.00975 |
| Sonnet 5.5 | $0.00019 | $0.00487 |
| Haiku 4.5 | $0.00010 | $0.00244 |
Grade A, and why
synthesize-decisions 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 2mo 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 — 242 lines — stays where its author put it; the contents beside it link to each section on GitHub.
When to use this skill
Promote design decisions from _analysis/design.md into normative practices — guidance that loads before a developer makes the wrong choice, not after they hit a bug. Complements /synthesize (Claude Code) or @synthesize (Cursor) (which promotes failure-mode practices from incident history and assembled cross-subsystem patterns).
Run this after /mine-design (Claude Code) or @mine-design (Cursor). Run /synthesize (Claude Code) or @synthesize (Cursor) separately for incident-history practices.
Step 1 — Read inputs
Read _analysis/design.md fully. If it doesn't exist, stop and tell the developer:
"
_analysis/design.mdnot found. Run/mine-design(Claude Code) or@mine-design(Cursor) first."
Read all sections: decisions, trade-offs, non-obvious behaviors, structural anti-patterns, and the ## Architectural patterns section. Record every finding.
Also list all directories under _analysis/extracted_skills_codebase/. Read the description field from each existing SKILL.md. Any decision already covered by an existing practice is skipped — do not duplicate.
Step 2 — Apply 7 concern lenses
Re-read every design.md finding through each of these 7 lenses. A finding can be relevant to more than one lens. Collect findings into concern clusters — you are grouping by class of problem, not by subsystem.
The probes 7–9 findings from mine-design are especially relevant here (explicit non-defaults, framework departures, deliberate failure prevention) — start with those.
Concern 1: Shared state and isolation
Where state that should be scoped to one request, coroutine, or session accidentally escapes into a wider scope.
Look for: process-global state set per-request, shared mutable objects accessed concurrently, missing per-coroutine isolation in async frameworks, global registries modified at runtime.
Framing question: What breaks under concurrent load that single-threaded testing never reveals?
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.
- 2mo ago First seen · 242 lines · 97 tokens per session scan A a1c2c34f5842
synthesize-decisions is a skill published in the GitHub repository ProsusAI/prism (20 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 97 tokens to every session and 2,437 once invoked, about $0.0004 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.
synthesize-decisions compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| synthesize-decisionsProsusAI/prism | Skill | 20 | 4mo ago | Apache-2.0 | A | 97 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |