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.
git clone --depth 1 https://github.com/Ronkupper/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/commands/ronkupper/prism/prism-meta)<a href="https://agentmods.dev/commands/ronkupper/prism/prism-meta"><img src="https://agentmods.dev/badge/commands/ronkupper/prism/prism-meta/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/commands/ronkupper/prism/prism-meta"><img src="https://agentmods.dev/badge/commands/ronkupper/prism/prism-meta.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 | $0.00037 | $0.00605 |
| Opus 5.5 | $0.00015 | $0.00242 |
| Sonnet 5.5 | $0.00007 | $0.00121 |
| Haiku 4.5 | $0.00004 | $0.00060 |
Grade A, and why
prism-meta 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.
What it actually says
Resume the PRISM Meta lane (the methodology lane — reflection, synthesis, worksheet-building) from its continuous repo state.
Optional meta pointer or focus note (may be empty):
$ARGUMENTS
Then proceed:
- Pull canonical repo state. The Meta lane is
repo_backed— work from a fresh clone orpull --rebaseof the live tail, never a stale checkout (commit-only-your-own-files; seePRISM_core.md§3.7.3). If the repo is not reachable, degrade to the pointer/handoff the operator supplies above. - Load the framework if it is not already active: read
PRISM_core.md(naming PRISM also auto-triggers theprismSkill, which loads the same core). - Adopt the session identity. Read the meta resume-pointer (the latest
meta handoff / meta What's next) and adopt the
Meta-<N>number it assigns — do not self-grab a number. Announce it ("Meta-N, resuming from ML-…"). - Drain the meta inbox. Read
OPEN_ITEMS_meta.md; fold each non-terminalcross-lane/ open item into the meta-log (promote any candidate-ML to a realML-Nfrom the live tail), then append adraineddisposition line. Draining is append-only — never edit a prior item in place. - Run the SP-13 substrate self-check and emit your M5 self-band (this
standing session's own context band, read against
Meta-N). Hand off proactively at a seam if your self-band warrants. - Proceed from the meta pointer (or the focus above, weighed against it); honor lane boundaries — append to the meta-log and update the meta pointer; route object-lane items to the object lane's inbox rather than acting on them here.
This command is a Claude-Skill convenience for entering the Meta lane; it
changes no framework mechanic. It is the contributor / methodology command —
the sibling of the engagement commands (/prism-start, /prism-whats-next,
/prism-converge, /prism-status, /prism-close), serving the lane that
develops the methodology itself rather than running an audit. The plain-language
form ("resume the PRISM meta lane") is equivalent and is the portable form on
every other vendor.
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 · 46 lines · 37 tokens per session scan A 6a086264db36
prism-meta is a command published in the GitHub repository Ronkupper/PRISM (5 stars, last pushed 4mo ago), licensed MIT. It adds 37 tokens to every session and 605 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 2026-08-31.
prism-meta compared with commands from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| prism-metaRonkupper/PRISM | Command | 5 | 4mo ago | MIT | A | 37 | Claude Code |
| renameasgeirtj/system_prompts_leaks | Command | 71k | 2d ago | CC0-1.0 | A | 0 | Claude Code |
| vibe-checkOneWave-AI/open-agent-stack | Command | 5 | 2mo ago | MIT | A | 20 | Claude Code |
| openai-logincezaraf/claude-code-gpt-pro | Command | 1 | 6mo ago | none | A | 17 | Claude Code |
| openai-offcezaraf/claude-code-gpt-pro | Command | 1 | 6mo ago | none | A | 11 | Claude Code |
| openai-statuscezaraf/claude-code-gpt-pro | Command | 1 | 6mo ago | none | A | 15 | Claude Code |
| checklistgithub/spec-kit | Command | 148k | yesterday | MIT | A | 13 | Claude Code |